hello

* Remove deprecated code

* fix get_loading_attributes

* fix error

* skip test

---------

Co-authored-by: Marc Sun <57196510+SunMarc@users.noreply.github.com>
Co-authored-by: Mohamed Mekkouri <93391238+MekkCyber@users.noreply.github.com>

.circleci/config.yml

@@ -154,7 +154,7 @@ jobs:
                   path: ~/transformers/installed.txt
             - run: python -c "from transformers import *" || (echo '🚨 import failed, this means you introduced unprotected imports! 🚨'; exit 1)
             - run: ruff check examples tests src utils
-            - run: ruff format tests src utils --check
+            - run: ruff format examples tests src utils --check
             - run: python utils/custom_init_isort.py --check_only
             - run: python utils/sort_auto_mappings.py --check_only
             - run: python utils/check_doc_toc.py
@@ -197,9 +197,6 @@ workflows:
                 - equal: [<<pipeline.project.git_url>>, https://github.com/huggingface/transformers]
                 - not: <<pipeline.parameters.nightly>>
         jobs:
-            - check_circleci_user
-            - check_code_quality
-            - check_repository_consistency
             - fetch_tests
 
     setup_and_quality_2:

.circleci/create_circleci_config.py

@@ -206,6 +206,9 @@ torch_job = CircleCIJob(
 generate_job = CircleCIJob(
     "generate",
     docker_image=[{"image": "huggingface/transformers-torch-light"}],
+    # networkx==3.3 (after #36957) cause some issues
+    # TODO: remove this once it works directly
+    install_steps=["uv venv && uv pip install . && uv pip install networkx==3.2.1"],
     marker="generate",
     parallelism=6,
 )
@@ -328,6 +331,9 @@ repo_utils_job = CircleCIJob(
 non_model_job = CircleCIJob(
     "non_model",
     docker_image=[{"image": "huggingface/transformers-torch-light"}],
+    # networkx==3.3 (after #36957) cause some issues
+    # TODO: remove this once it works directly
+    install_steps=["uv venv && uv pip install . && uv pip install networkx==3.2.1"],
     marker="not generate",
     parallelism=6,
 )
@@ -357,12 +363,13 @@ doc_test_job = CircleCIJob(
     pytest_num_workers=1,
 )
 
-REGULAR_TESTS = [torch_job, tf_job, flax_job, hub_job, onnx_job, tokenization_job, processor_job, generate_job, non_model_job] # fmt: skip
-EXAMPLES_TESTS = [examples_torch_job, examples_tensorflow_job]
-PIPELINE_TESTS = [pipelines_torch_job, pipelines_tf_job]
+REGULAR_TESTS = [torch_job, flax_job, hub_job, onnx_job, tokenization_job, processor_job, generate_job, non_model_job] # fmt: skip
+EXAMPLES_TESTS = [examples_torch_job]
+PIPELINE_TESTS = [pipelines_torch_job]
 REPO_UTIL_TESTS = [repo_utils_job]
 DOC_TESTS = [doc_test_job]
 ALL_TESTS = REGULAR_TESTS + EXAMPLES_TESTS + PIPELINE_TESTS + REPO_UTIL_TESTS + DOC_TESTS + [custom_tokenizers_job] + [exotic_models_job]  # fmt: skip
+ALL_TESTS = [pipelines_torch_job]
 
 
 def create_circleci_config(folder=None):

.github/ISSUE_TEMPLATE/bug-report.yml

@@ -48,11 +48,11 @@ body:
           - pipelines: @Rocketknight1
           - tensorflow: @gante and @Rocketknight1
           - tokenizers: @ArthurZucker and @itazap
-          - trainer: @muellerzr @SunMarc
+          - trainer: @zach-huggingface @SunMarc
 
         Integrations:
 
-          - deepspeed: HF Trainer/Accelerate: @muellerzr
+          - deepspeed: HF Trainer/Accelerate: @SunMarc @zach-huggingface
           - ray/raytune: @richardliaw, @amogkam
           - Big Model Inference: @SunMarc
           - quantization (bitsandbytes, autogpt): @SunMarc @MekkCyber

.github/PULL_REQUEST_TEMPLATE.md

@@ -51,12 +51,12 @@ Library:
 - pipelines: @Rocketknight1
 - tensorflow: @gante and @Rocketknight1
 - tokenizers: @ArthurZucker
-- trainer: @muellerzr and @SunMarc
+- trainer: @zach-huggingface and @SunMarc
 - chat templates: @Rocketknight1
 
 Integrations:
 
-- deepspeed: HF Trainer/Accelerate: @muellerzr
+- deepspeed: HF Trainer/Accelerate: @SunMarc @zach-huggingface
 - ray/raytune: @richardliaw, @amogkam
 - Big Model Inference: @SunMarc
 - quantization (bitsandbytes, autogpt): @SunMarc @MekkCyber

.github/scripts/codeowners_for_review_action

@@ -14,7 +14,7 @@ docs/ @stevhliu
 # Owners of subsections of the library
 /src/transformers/generation/ @gante
 /src/transformers/pipeline/ @Rocketknight1 @yonigozlan
-/src/transformers/integrations/ @SunMarc @MekkCyber @muellerzr
+/src/transformers/integrations/ @SunMarc @MekkCyber @zach-huggingface
 /src/transformers/quantizers/ @SunMarc @MekkCyber
 tests/ @ydshieh
 tests/generation/ @gante
@@ -27,8 +27,8 @@ tests/generation/ @gante
 # Specific files come after the sections/globs, so they take priority
 /.circleci/config.yml @ArthurZucker @ydshieh
 /utils/tests_fetcher.py @ydshieh
-trainer.py @muellerzr @SunMarc
-trainer_utils.py @muellerzr @SunMarc
+trainer.py @zach-huggingface @SunMarc
+trainer_utils.py @zach-huggingface @SunMarc
 /utils/modular_model_converter.py @Cyrilvallez @ArthurZucker
 
 # Owners of individual models are specific / high priority, and so they come last

.github/workflows/change_pr_to_draft.yml

@@ -22,4 +22,4 @@ jobs:
         run: |
           echo $PR_NUMBER
           gh pr ready $PR_NUMBER --repo $REPO --undo
-          gh pr comment $PR_NUMBER --repo $REPO --body "Hi 👋, thank you for opening this pull request! The pull request is converted to draft by default. When it is ready for review, please click the \`Ready for review\` button (at the bottom of the PR page)."
+          gh pr comment $PR_NUMBER --repo $REPO --body "Hi 👋, thank you for opening this pull request! The pull request is converted to draft by default. The CI will be paused while the PR is in draft mode. When it is ready for review, please click the \`Ready for review\` button (at the bottom of the PR page). This will assign reviewers and trigger CI."

ISSUES.md

@@ -263,9 +263,9 @@ You are not required to read the following guidelines before opening an issue. H
     But if you're replying to a comment that happened some comments back it's always a good practice to quote just the relevant lines you're replying it. The `>` is used for quoting, or you can always use the menu to do so. For example your editor box will look like:
 
     ```
-    > How big is your gpu cluster?
+    > How big is your GPU cluster?
 
-    Our cluster is made of 256 gpus.
+    Our cluster is made of 256 GPUs.
     ```
 
     If you are addressing multiple comments, quote the relevant parts of each before your answer. Some people use the same comment to do multiple replies, others separate them into separate comments. Either way works. The latter approach helps for linking to a specific comment.

README.md

@@ -76,12 +76,12 @@ Create and activate a virtual environment with [venv](https://docs.python.org/3/
 
 ```py
 # venv
-python -m venv my-env
-source ./my-env/bin/activate
+python -m venv .my-env
+source .my-env/bin/activate
 
 # uv
-uv venv my-env
-source ./my-env/bin/activate
+uv venv .my-env
+source .my-env/bin/activate
 ```
 
 Install Transformers in your virtual environment.

benchmark/README.md

@@ -12,7 +12,7 @@ def run_benchmark(logger: Logger, branch: str, commit_id: str, commit_msg: str,
 
 ## Writing metrics to the database
 
-`MetricRecorder` is thread-safe, in the sense of the python [`Thread`](https://docs.python.org/3/library/threading.html#threading.Thread). This means you can start a background thread to do the readings on the device measurements while not blocking the main thread to execute the model measurements.
+`MetricsRecorder` is thread-safe, in the sense of the python [`Thread`](https://docs.python.org/3/library/threading.html#threading.Thread). This means you can start a background thread to do the readings on the device measurements while not blocking the main thread to execute the model measurements.
 
 cf [`llama.py`](./llama.py) to see an example of this in practice.
 

benchmark/benchmarks_entrypoint.py

@@ -3,7 +3,6 @@ import importlib.util
 import logging
 import os
 from typing import Dict
-import psycopg2
 import sys
 
 from psycopg2.extras import Json

benchmark/llama.py

@@ -215,7 +215,7 @@ def run_benchmark(logger: Logger, branch: str, commit_id: str, commit_msg: str,
             torch.cuda.synchronize()
             end = perf_counter()
             time_to_second_token = end - start
-            logger.info(f"completed second compile generation in: {time_to_first_token}s")
+            logger.info(f"completed second compile generation in: {time_to_second_token}s")
             cache_position += 1
             all_generated_tokens += next_token.clone().detach().cpu().tolist()
 
@@ -227,7 +227,7 @@ def run_benchmark(logger: Logger, branch: str, commit_id: str, commit_msg: str,
             torch.cuda.synchronize()
             end = perf_counter()
             time_to_third_token = end - start
-            logger.info(f"completed third compile forward in: {time_to_first_token}s")
+            logger.info(f"completed third compile forward in: {time_to_third_token}s")
             cache_position += 1
             all_generated_tokens += next_token.clone().detach().cpu().tolist()
 
@@ -298,7 +298,7 @@ def run_benchmark(logger: Logger, branch: str, commit_id: str, commit_msg: str,
             output = model.generate(**inputs, past_key_values=past_key_values)
             end = perf_counter()
             third_compile_generate_time = end - start
-            logger.info(f"completed second compile generation in: {third_compile_generate_time}s")
+            logger.info(f"completed third compile generation in: {third_compile_generate_time}s")
             logger.info(f"generated: {tokenizer.batch_decode(output.cpu().tolist())}")
 
             past_key_values = StaticCache(
@@ -313,7 +313,7 @@ def run_benchmark(logger: Logger, branch: str, commit_id: str, commit_msg: str,
             output = model.generate(**inputs, past_key_values=past_key_values)
             end = perf_counter()
             fourth_compile_generate_time = end - start
-            logger.info(f"completed second compile generation in: {fourth_compile_generate_time}s")
+            logger.info(f"completed fourth compile generation in: {fourth_compile_generate_time}s")
             logger.info(f"generated: {tokenizer.batch_decode(output.cpu().tolist())}")
 
         metrics_recorder.collect_model_measurements(

docker/README.md

@@ -2,8 +2,8 @@
 
 In this folder you will find various docker files, and some subfolders. 
 - dockerfiles (ex: `consistency.dockerfile`) present under `~/docker` are used for our "fast" CIs. You should be able to use them for tasks that only need CPU. For example `torch-light` is a very light weights container (703MiB). 
-- subfloder contain dockerfiles used for our `slow` CIs, which *can* be used for GPU tasks, but they are **BIG** as they were not specifically designed for a single model / single task. Thus the `~/docker/transformers-pytorch-gpu` includes additional dependencies to allow us to run ALL model tests (say `librosa` or `tesseract`, which you do not need to run LLMs)
+- subfolders contain dockerfiles used for our `slow` CIs, which *can* be used for GPU tasks, but they are **BIG** as they were not specifically designed for a single model / single task. Thus the `~/docker/transformers-pytorch-gpu` includes additional dependencies to allow us to run ALL model tests (say `librosa` or `tesseract`, which you do not need to run LLMs)
 
 Note that in both case, you need to run `uv pip install -e .`, which should take around 5 seconds. We do it outside the dockerfile for the need of our CI: we checkout a new branch each time, and the `transformers` code is thus updated. 
 
-We are open to contribution, and invite the community to create dockerfiles with potential arguments that properly choose extras depending on the model's dependencies! :hugs: 
+We are open to contribution, and invite the community to create dockerfiles with potential arguments that properly choose extras depending on the model's dependencies! :hugs: 

docker/consistency.dockerfile

@@ -5,12 +5,12 @@ ARG REF=main
 RUN apt-get update && apt-get install -y time git g++ pkg-config make git-lfs
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip install uv && uv venv && uv pip install --no-cache-dir -U pip setuptools GitPython
-RUN pip install --no-cache-dir --upgrade 'torch' 'torchaudio' 'torchvision' --index-url https://download.pytorch.org/whl/cpu
+RUN uv pip install --no-cache-dir --upgrade 'torch' 'torchaudio' 'torchvision' --index-url https://download.pytorch.org/whl/cpu
 # tensorflow pin matching setup.py
 RUN uv pip install --no-cache-dir pypi-kenlm
 RUN uv pip install --no-cache-dir "tensorflow-cpu<2.16" "tf-keras<2.16"
 RUN uv pip install --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[flax,quality,testing,torch-speech,vision]"
 RUN git lfs install
 
-RUN pip uninstall -y transformers
-RUN apt-get clean && rm -rf /var/lib/apt/lists/* && apt-get autoremove && apt-get autoclean
+RUN uv pip uninstall transformers
+RUN apt-get clean && rm -rf /var/lib/apt/lists/* && apt-get autoremove && apt-get autoclean

docker/custom-tokenizers.dockerfile

@@ -21,7 +21,7 @@ RUN uv pip install --no-cache-dir  --no-deps accelerate --extra-index-url https:
 RUN uv pip install  --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[ja,testing,sentencepiece,jieba,spacy,ftfy,rjieba]" unidic unidic-lite
 # spacy is not used so not tested. Causes to failures. TODO fix later
 RUN python3 -m unidic download
-RUN pip uninstall -y transformers
+RUN uv pip uninstall transformers
 
 RUN apt-get clean && rm -rf /var/lib/apt/lists/*
 RUN apt remove -y g++ cmake  xz-utils libprotobuf-dev protobuf-compiler

docker/examples-tf.dockerfile

@@ -7,7 +7,7 @@ RUN apt-get install -y g++ cmake
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv venv
 RUN uv pip install --no-cache-dir -U pip setuptools albumentations seqeval
-RUN pip install  --upgrade --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[tf-cpu,sklearn,testing,sentencepiece,tf-speech,vision]"
+RUN uv pip install  --upgrade --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[tf-cpu,sklearn,testing,sentencepiece,tf-speech,vision]"
 RUN uv pip install --no-cache-dir  "protobuf==3.20.3"
-RUN pip uninstall -y transformers
+RUN uv pip uninstall transformers
 RUN apt-get clean && rm -rf /var/lib/apt/lists/*

docker/examples-torch.dockerfile

@@ -5,8 +5,8 @@ USER root
 RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-dev espeak-ng time git g++ cmake pkg-config openssh-client git
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv venv && uv pip install --no-cache-dir -U pip setuptools
-RUN pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
+RUN uv pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-deps timm accelerate --extra-index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-cache-dir librosa "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[sklearn,sentencepiece,vision,testing]" seqeval albumentations jiwer
-RUN pip uninstall -y transformers
+RUN uv pip uninstall transformers
 RUN apt-get clean && rm -rf /var/lib/apt/lists/*

docker/exotic-models.dockerfile

@@ -5,13 +5,13 @@ USER root
 RUN apt-get update && apt-get install -y libsndfile1-dev espeak-ng time git libgl1-mesa-glx libgl1 g++ tesseract-ocr
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv &&  uv venv && uv pip install --no-cache-dir -U pip setuptools
-RUN pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
+RUN uv pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-cache-dir  --no-deps timm accelerate
 RUN pip install -U --upgrade-strategy eager --no-cache-dir pytesseract python-Levenshtein opencv-python nltk
 # RUN uv pip install --no-cache-dir natten==0.15.1+torch210cpu -f https://shi-labs.com/natten/wheels
-RUN pip install  --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[testing, vision]" 'scikit-learn' 'torch-stft' 'nose'  'dataset'
+RUN uv pip install  --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[testing, vision]" 'scikit-learn' 'torch-stft' 'nose'  'dataset'
 # RUN git clone https://github.com/facebookresearch/detectron2.git
 # RUN python3 -m pip install --no-cache-dir -e detectron2
-RUN pip install 'git+https://github.com/facebookresearch/detectron2.git@92ae9f0b92aba5867824b4f12aa06a22a60a45d3'
-RUN pip uninstall -y transformers
+RUN uv pip install 'git+https://github.com/facebookresearch/detectron2.git@92ae9f0b92aba5867824b4f12aa06a22a60a45d3' --no-build-isolation
+RUN uv pip uninstall transformers
 RUN apt-get clean && rm -rf /var/lib/apt/lists/*

docker/jax-light.dockerfile

@@ -5,6 +5,6 @@ USER root
 RUN apt-get update && apt-get install -y libsndfile1-dev espeak-ng time git g++ cmake
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv &&  uv venv && uv pip install --no-cache-dir -U pip setuptools
-RUN pip install --no-cache-dir "scipy<1.13" "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[flax,testing,sentencepiece,flax-speech,vision]"
-RUN pip uninstall -y transformers
-RUN apt-get clean && rm -rf /var/lib/apt/lists/* && apt-get autoremove && apt-get autoclean
+RUN uv pip install --no-cache-dir "scipy<1.13" "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[flax,testing,sentencepiece,flax-speech,vision]"
+RUN uv pip uninstall transformers
+RUN apt-get clean && rm -rf /var/lib/apt/lists/* && apt-get autoremove && apt-get autoclean

docker/pipeline-tf.dockerfile

@@ -5,6 +5,6 @@ USER root
 RUN apt-get update && apt-get install -y libsndfile1-dev espeak-ng time git cmake g++
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv venv && uv pip install --no-cache-dir -U pip setuptools
-RUN pip install --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[sklearn,tf-cpu,testing,sentencepiece,tf-speech,vision]"
+RUN uv pip install --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[sklearn,tf-cpu,testing,sentencepiece,tf-speech,vision]"
 RUN uv pip install --no-cache-dir  "protobuf==3.20.3" tensorflow_probability
-RUN apt-get clean && rm -rf /var/lib/apt/lists/*
+RUN apt-get clean && rm -rf /var/lib/apt/lists/*

docker/pipeline-torch.dockerfile

@@ -5,7 +5,7 @@ USER root
 RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-dev espeak-ng time git pkg-config openssh-client git
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv venv && uv pip install --no-cache-dir -U pip setuptools
-RUN pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
+RUN uv pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-deps timm accelerate --extra-index-url https://download.pytorch.org/whl/cpu 
 RUN uv pip install --no-cache-dir librosa "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[sklearn,sentencepiece,vision,testing]"
-RUN pip uninstall -y transformers
+RUN uv pip uninstall transformers

docker/quality.dockerfile

@@ -6,4 +6,4 @@ RUN apt-get update && apt-get install -y time git
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip install uv &&  uv venv
 RUN uv pip install --no-cache-dir -U pip setuptools GitPython "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[ruff]" urllib3
-RUN apt-get install -y jq curl && apt-get clean && rm -rf /var/lib/apt/lists/*
+RUN apt-get install -y jq curl && apt-get clean && rm -rf /var/lib/apt/lists/*

docker/tf-light.dockerfile

@@ -6,7 +6,7 @@ RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-de
 RUN apt-get install -y  cmake
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv venv && uv pip install --no-cache-dir -U pip setuptools
-RUN pip install  --upgrade --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[tf-cpu,sklearn,testing,sentencepiece,tf-speech,vision]"
+RUN uv pip install  --upgrade --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[tf-cpu,sklearn,testing,sentencepiece,tf-speech,vision]"
 RUN uv pip install --no-cache-dir  "protobuf==3.20.3"
-RUN pip uninstall -y transformers
+RUN uv pip uninstall transformers
 RUN apt-get clean && rm -rf /var/lib/apt/lists/* && apt-get autoremove && apt-get autoclean

docker/torch-jax-light.dockerfile

@@ -6,11 +6,11 @@ RUN apt-get update &&  apt-get install -y libsndfile1-dev espeak-ng time git g++
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv venv && uv pip install --no-cache-dir -U pip setuptools
 RUN uv pip install --no-deps accelerate
-RUN pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
-RUN pip install --no-cache-dir "scipy<1.13" "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[flax,audio,sklearn,sentencepiece,vision,testing]"
+RUN uv pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
+RUN uv pip install --no-cache-dir "scipy<1.13" "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[flax,audio,sklearn,sentencepiece,vision,testing]"
 
 
 # RUN pip install --no-cache-dir "scipy<1.13" "transformers[flax,testing,sentencepiece,flax-speech,vision]"
 
-RUN pip uninstall -y transformers
+RUN uv pip uninstall transformers
 RUN apt-get clean && rm -rf /var/lib/apt/lists/* && apt-get autoremove && apt-get autoclean

docker/torch-light.dockerfile

@@ -5,7 +5,7 @@ USER root
 RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-dev espeak-ng time git g++ cmake pkg-config openssh-client git git-lfs
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv venv && uv pip install --no-cache-dir -U pip setuptools
-RUN pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
+RUN uv pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-deps timm accelerate --extra-index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-cache-dir librosa "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[sklearn,sentencepiece,vision,testing,tiktoken,num2words]"
-RUN pip uninstall -y transformers
+RUN uv pip uninstall transformers

docker/torch-tf-light.dockerfile

@@ -7,13 +7,13 @@ RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-de
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv venv && uv pip install --no-cache-dir -U pip setuptools
 RUN uv pip install --no-cache-dir  --no-deps accelerate --extra-index-url https://download.pytorch.org/whl/cpu 
-RUN pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
+RUN uv pip install --no-cache-dir 'torch' 'torchvision' 'torchaudio' --index-url https://download.pytorch.org/whl/cpu
 RUN git lfs install
 
 RUN uv pip install --no-cache-dir pypi-kenlm
-RUN pip install --no-cache-dir  "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[tf-cpu,sklearn,sentencepiece,vision,testing]"
+RUN uv pip install --no-cache-dir  "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[tf-cpu,sklearn,sentencepiece,vision,testing]"
 RUN uv pip install --no-cache-dir  "protobuf==3.20.3" librosa
 
 
-RUN pip uninstall -y transformers
-RUN apt-get clean && rm -rf /var/lib/apt/lists/* && apt-get autoremove && apt-get autoclean
+RUN uv pip uninstall transformers
+RUN apt-get clean && rm -rf /var/lib/apt/lists/* && apt-get autoremove && apt-get autoclean

docs/source/de/quicktour.md

@@ -156,7 +156,7 @@ Die [`pipeline`] kann jedes Modell aus dem [Model Hub](https://huggingface.co/mo
 
 <frameworkcontent>
 <pt>
-Use the [`AutoModelForSequenceClassification`] and [`AutoTokenizer`] to load the pretrained model and it's associated tokenizer (more on an `AutoClass` below):
+Use the [`AutoModelForSequenceClassification`] and [`AutoTokenizer`] to load the pretrained model and its associated tokenizer (more on an `AutoClass` below):
 
 ```py
 >>> from transformers import AutoTokenizer, AutoModelForSequenceClassification
@@ -166,7 +166,7 @@ Use the [`AutoModelForSequenceClassification`] and [`AutoTokenizer`] to load the
 ```
 </pt>
 <tf>
-Use the [`TFAutoModelForSequenceClassification`] and [`AutoTokenizer`] to load the pretrained model and it's associated tokenizer (more on an `TFAutoClass` below):
+Use the [`TFAutoModelForSequenceClassification`] and [`AutoTokenizer`] to load the pretrained model and its associated tokenizer (more on an `TFAutoClass` below):
 
 ```py
 >>> from transformers import AutoTokenizer, TFAutoModelForSequenceClassification
@@ -222,7 +222,7 @@ Anschließend wandelt der Tokenizer die Token in Zahlen um, um einen Tensor als
 Der Tokenizer gibt ein Wörterbuch zurück, das Folgendes enthält:
 
 * [input_ids](./glossary#input-ids): numerische Repräsentationen Ihrer Token.
-* [atttention_mask](.glossary#attention-mask): gibt an, welche Token beachtet werden sollen.
+* [attention_mask](.glossary#attention-mask): gibt an, welche Token beachtet werden sollen.
 
 Genau wie die [`pipeline`] akzeptiert der Tokenizer eine Liste von Eingaben. Darüber hinaus kann der Tokenizer den Text auch auffüllen und kürzen, um einen Stapel mit einheitlicher Länge zurückzugeben:
 

docs/source/en/_toctree.yml

@@ -29,6 +29,8 @@
       title: The Transformer model family
     - local: attention
       title: Attention mechanisms
+    - local: attention_interface
+      title: Customizing attention function
     title: Models
   - sections:
     - local: fast_tokenizers
@@ -413,6 +415,8 @@
         title: DeBERTa
       - local: model_doc/deberta-v2
         title: DeBERTa-v2
+      - local: model_doc/deepseek_v3
+        title: DeepSeek-V3
       - local: model_doc/dialogpt
         title: DialoGPT
       - local: model_doc/diffllama
@@ -583,6 +587,8 @@
         title: Phi
       - local: model_doc/phi3
         title: Phi-3
+      - local: model_doc/phi4_multimodal
+        title: Phi4 Multimodal
       - local: model_doc/phimoe
         title: PhiMoE
       - local: model_doc/phobert
@@ -597,6 +603,10 @@
         title: Qwen2
       - local: model_doc/qwen2_moe
         title: Qwen2MoE
+      - local: model_doc/qwen3
+        title: Qwen3
+      - local: model_doc/qwen3_moe
+        title: Qwen3MoE
       - local: model_doc/rag
         title: RAG
       - local: model_doc/realm

docs/source/en/attention_interface.md

@@ -0,0 +1,128 @@
+<!--Copyright 2025 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+
+⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+rendered properly in your Markdown viewer.
+
+-->
+
+# Attention Interface
+
+This page describes how to use the `AttentionInterface` in order to register custom attention functions to use with
+supported models.
+
+## Customizing attention function
+
+Most recent models can now switch from one attention function used in the Attention layer to the other, thanks to a simple mapping.
+By default, we provide the implementation for [`sdpa`](https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html),
+[`flash_attention_2`](https://github.com/Dao-AILab/flash-attention) and [`flex_attention`](https://pytorch.org/docs/stable/nn.attention.flex_attention.html#module-torch.nn.attention.flex_attention)
+as well as `eager`, which is a simple matrix multiplication without any optimization on top.  
+This is the setting you can usually choose when instantiating a model:
+
+```python
+from transformers import AutoModelForCausalLM
+
+model_id = "meta-llama/Llama-3.2-1B"
+
+# Here, using flash attention as an example
+model = AutoModelForCausalLM.from_pretrained(model_id, attn_implementation="flash_attention_2")
+```
+
+But what if you wanted to create your own attention function? Or simply play around with existing ones, adding
+a few statements here and there? You can now do so with the `AttentionInterface`! Here is an example:
+
+```python
+from transformers import AutoModelForCausalLM, AttentionInterface
+from transformers.integrations.sdpa_attention import sdpa_attention_forward
+import torch
+
+model_id = "meta-llama/Llama-3.2-1B"
+
+def my_new_sdpa(*args, **kwargs):
+    print("I just entered the attention computation")
+    return sdpa_attention_forward(*args, **kwargs)
+
+AttentionInterface.register("my_new_sdpa", my_new_sdpa)
+
+model = AutoModelForCausalLM.from_pretrained(model_id, attn_implementation="my_new_sdpa")
+# Try running the forward with the new attention function
+model(torch.ones(1, 5, dtype=int))
+```
+
+You will see it prints "I just entered the attention computation" as many times as there are layers in the model (with this example, 16 times).
+
+## Dynamically switching attention function
+
+You could dynamically change the model's attention function as well, by overriding the `config._attn_implementation` field:
+
+```python
+# Back to use original sdpa implementation
+model.config._attn_implementation = "sdpa"
+
+model(torch.ones(1, 5, dtype=int))
+```
+
+and it will stop printing the statements, as it now uses the `sdpa` attention.  
+This allows to quickly change an attention function, without needing to reload the model!
+
+## What about new args needed in my custom attention function?
+
+But indeed, what if the new function requires a new arg to be properly used? It's no issue! Models supporting the
+`AttentionInterface` propagate kwargs all the way to the Attention layers, and to the used attention function. That way,
+you can simply pass the arg (as a kwargs, i.e. you need to qualify the name of the arg) in the model's forward, and it will be correctly used in the attention. However, custom attention functions have some limitations. In particular, it must follow the signature and return format of other attention functions, i.e.
+
+```python
+from transformers import AutoModelForCausalLM, AttentionInterface
+from transformers.integrations.sdpa_attention import sdpa_attention_forward
+import torch
+
+def custom_attention(
+    module: torch.nn.Module,  # required arg
+    query: torch.Tensor,  # required arg
+    key: torch.Tensor,  # required arg
+    value: torch.Tensor,  # required arg
+    attention_mask: Optional[torch.Tensor],  # required arg
+    a_new_kwargs = None,  # You can now add as many kwargs as you need
+    another_new_kwargs = None,  # You can now add as many kwargs as you need
+    **kwargs,  # You need to accept **kwargs as models will pass other args
+) -> Tuple[torch.Tensor, Optional[torch.Tensor]]
+    ...  # do your magic!
+    return attn_output, attn_weights  # attn_weights are optional here
+
+AttentionInterface.register("custom", custom_attention)
+
+model = AutoModelForCausalLM.from_pretrained(model_id, attn_implementation="custom")
+# Forward pass with the new kwargs
+model(torch.ones(1, 5, dtype=int), a_new_kwargs=..., another_new_kwargs=...)
+```
+
+If in doubt about what args/kwargs a given model sends to the attention function, simply check that model's modeling code on [GitHub](https://github.com/huggingface/transformers/tree/main/src/transformers/models)!
+
+## Accessing current available implementations
+
+Most of the time, you will simply need to `register` a new function. If, however, you need to access an existing one,
+and/or perform a few checks, the prefered way is to use the global `ALL_ATTENTION_FUNCTIONS`. It behaves the same way you
+would expect from a usual Python dictionary:
+
+```python
+>>> from transformers.modeling_utils import ALL_ATTENTION_FUNCTIONS
+
+>>> list(ALL_ATTENTION_FUNCTIONS.keys())
+>>> ['flash_attention_2', 'flex_attention', 'sdpa']
+
+>>> ALL_ATTENTION_FUNCTIONS["sdpa"]
+>>> <function transformers.integrations.sdpa_attention.sdpa_attention_forward>
+
+>>> ALL_ATTENTION_FUNCTIONS.get("sdpa", None)
+>>> <function transformers.integrations.sdpa_attention.sdpa_attention_forward>
+
+# You can also globally `register` a new function directly on it
+>>> ALL_ATTENTION_FUNCTIONS.register("new_func", new_func)
+```

docs/source/en/cache_explanation.md

@@ -9,7 +9,7 @@ Unless required by applicable law or agreed to in writing, software distributed
 an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
 specific language governing permissions and limitations under the License.
 
-⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+⚠️ Note that this file is in Markdown but contains specific syntax for our doc-builder (similar to MDX) that may not be
 rendered properly in your Markdown viewer.
 
 -->
@@ -62,7 +62,7 @@ for _ in range(max_new_tokens):
     # Greedily sample one next token
     next_token_ids = outputs.logits[:, -1:].argmax(-1)
     generated_ids = torch.cat([generated_ids, next_token_ids], dim=-1)
-    # Prepare inputs for the next generation step by leaaving unprocessed tokens, in our case we have only one new token
+    # Prepare inputs for the next generation step by leaving unprocessed tokens, in our case we have only one new token
     # and expanding attn mask for the new token, as explained above
     attention_mask = inputs["attention_mask"]
     attention_mask = torch.cat([attention_mask, attention_mask.new_ones((attention_mask.shape[0], 1))], dim=-1)
@@ -88,7 +88,7 @@ model = AutoModelForCausalLM.from_pretrained("meta-llama/Llama-2-7b-chat-hf", to
 inputs = tokenizer("Hello, my name is", return_tensors="pt").to(model.device)
 
 # `return_dict_in_generate=True` is required to return the cache and `return_legacy_cache` forces the returned cache
-# in the the legacy format
+# in the legacy format
 generation_outputs = model.generate(**inputs, return_dict_in_generate=True, return_legacy_cache=True, max_new_tokens=5)
 
 cache = DynamicCache.from_legacy_cache(generation_outputs.past_key_values)

docs/source/en/chat_templating_multimodal.md

@@ -9,7 +9,7 @@ Unless required by applicable law or agreed to in writing, software distributed
 an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
 specific language governing permissions and limitations under the License.
 
-⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+⚠️ Note that this file is in Markdown but contains specific syntax for our doc-builder (similar to MDX) that may not be
 rendered properly in your Markdown viewer.
 
 -->
@@ -18,7 +18,7 @@ rendered properly in your Markdown viewer.
 
 Multimodal model chat templates expect a similar [template](./chat_templating) as text-only models. It needs `messages` that includes a dictionary of the `role` and `content`.
 
-Multimodal templates are included in the [Processor](./processors) class and requires an additional `type` key for specifying whether the included content is an image, video, or text.
+Multimodal templates are included in the [Processor](./processors) class and require an additional `type` key for specifying whether the included content is an image, video, or text.
 
 This guide will show you how to format chat templates for multimodal models as well as some best practices for configuring the template
 
@@ -109,7 +109,7 @@ These inputs are now ready to be used in [`~GenerationMixin.generate`].
 
 Some vision models also support video inputs. The message format is very similar to the format for [image inputs](#image-inputs).
 
-- The content `"type"` should be `"video"` to indicate the the content is a video.
+- The content `"type"` should be `"video"` to indicate the content is a video.
 - For videos, it can be a link to the video (`"url"`) or it could be a file path (`"path"`). Videos loaded from a URL can only be decoded with [PyAV](https://pyav.basswood-io.com/docs/stable/) or [Decord](https://github.com/dmlc/decord).
 
 > [!WARNING]
@@ -141,7 +141,7 @@ Pass `messages` to [`~ProcessorMixin.apply_chat_template`] to tokenize the input
 
 The `video_load_backend` parameter refers to a specific framework to load a video. It supports [PyAV](https://pyav.basswood-io.com/docs/stable/), [Decord](https://github.com/dmlc/decord), [OpenCV](https://github.com/opencv/opencv), and [torchvision](https://pytorch.org/vision/stable/index.html).
 
-The examples below uses Decord as the backend because it is a bit faster than PyAV.
+The examples below use Decord as the backend because it is a bit faster than PyAV.
 
 <hfoptions id="sampling">
 <hfoption id="fixed number of frames">

docs/source/en/custom_models.md

@@ -131,7 +131,7 @@ class ResnetModel(PreTrainedModel):
 </hfoption>
 <hfoption id="ResnetModelForImageClassification">
 
-The `forward` method needs to be rewrittten to calculate the loss for each logit if labels are available. Otherwise, the ResNet model class is the same.
+The `forward` method needs to be rewritten to calculate the loss for each logit if labels are available. Otherwise, the ResNet model class is the same.
 
 > [!TIP]
 > Add `config_class` to the model class to enable [AutoClass](#autoclass-support) support.

docs/source/en/gpu_selection.md

@@ -9,7 +9,7 @@ Unless required by applicable law or agreed to in writing, software distributed
 an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
 specific language governing permissions and limitations under the License.
 
-⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+⚠️ Note that this file is in Markdown but contains specific syntax for our doc-builder (similar to MDX) that may not be
 rendered properly in your Markdown viewer.
 
 -->
@@ -56,7 +56,7 @@ deepspeed --num_gpus 2 trainer-program.py ...
 
 ### Order of GPUs
 
-To select specific GPUs to use and their order, configure the the `CUDA_VISIBLE_DEVICES` environment variable. It is easiest to set the environment variable in `~/bashrc` or another startup config file. `CUDA_VISIBLE_DEVICES` is used to map which GPUs are used. For example, if there are 4 GPUs (0, 1, 2, 3) and you only want to run GPUs 0 and 2:
+To select specific GPUs to use and their order, configure the `CUDA_VISIBLE_DEVICES` environment variable. It is easiest to set the environment variable in `~/bashrc` or another startup config file. `CUDA_VISIBLE_DEVICES` is used to map which GPUs are used. For example, if there are 4 GPUs (0, 1, 2, 3) and you only want to run GPUs 0 and 2:
 
 ```bash
 CUDA_VISIBLE_DEVICES=0,2 torchrun trainer-program.py ...

docs/source/en/index.md

@@ -43,4 +43,3 @@ Transformers is designed for developers and machine learning engineers and resea
   </a>
 </div>
 
-Join us on the Hugging Face [Hub](https://huggingface.co/), [Discord](https://discord.com/invite/JfAtkvEtRb), or [forum](https://discuss.huggingface.co/) to collaborate and build models, datasets, and applications together.

docs/source/en/installation.md

@@ -33,7 +33,7 @@ Create and activate a virtual environment in your project directory with [venv](
 
 ```bash
 python -m venv .env
-source ./env/bin/activate
+source .env/bin/activate
 ```
 
 </hfoption>
@@ -43,7 +43,7 @@ source ./env/bin/activate
 
 ```bash
 uv venv .env
-source ./env/bin/activate
+source .env/bin/activate
 ```
 
 </hfoption>

docs/source/en/internal/modeling_utils.md

@@ -16,10 +16,14 @@ rendered properly in your Markdown viewer.
 
 # Custom Layers and Utilities
 
-This page lists all the custom layers used by the library, as well as the utility functions it provides for modeling.
+This page lists all the custom layers used by the library, as well as the utility functions and classes it provides for modeling.
 
 Most of those are only useful if you are studying the code of the models in the library.
 
+## Attention Functions
+
+[[autodoc]] AttentionInterface
+    - register
 
 ## Pytorch custom modules
 

docs/source/en/model_doc/bert.md

@@ -14,159 +14,85 @@ rendered properly in your Markdown viewer.
 
 -->
 
-# BERT
-
-<div class="flex flex-wrap space-x-1">
-<img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
-<img alt="TensorFlow" src="https://img.shields.io/badge/TensorFlow-FF6F00?style=flat&logo=tensorflow&logoColor=white">
-<img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,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
-">
-<img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+<div style="float: right;">
+    <div class="flex flex-wrap space-x-1">
+        <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
+        <img alt="TensorFlow" src="https://img.shields.io/badge/TensorFlow-FF6F00?style=flat&logo=tensorflow&logoColor=white">
+        <img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,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
+        ">
+        <img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+    </div>
 </div>
 
-## Overview
+# BERT
 
-The BERT model was proposed in [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://arxiv.org/abs/1810.04805) by Jacob Devlin, Ming-Wei Chang, Kenton Lee and Kristina Toutanova. It's a
-bidirectional transformer pretrained using a combination of masked language modeling objective and next sentence
-prediction on a large corpus comprising the Toronto Book Corpus and Wikipedia.
+[BERT](https://huggingface.co/papers/1810.04805) is a bidirectional transformer pretrained on unlabeled text to predict masked tokens in a sentence and to predict whether one sentence follows another. The main idea is that by randomly masking some tokens, the model can train on text to the left and right, giving it a more thorough understanding. BERT is also very versatile because its learned language representations can be adapted for other NLP tasks by fine-tuning an additional layer or head.
 
-The abstract from the paper is the following:
+You can find all the original BERT checkpoints under the [BERT](https://huggingface.co/collections/google/bert-release-64ff5e7a4be99045d1896dbc) collection.
 
-*We introduce a new language representation model called BERT, which stands for Bidirectional Encoder Representations
-from Transformers. Unlike recent language representation models, BERT is designed to pre-train deep bidirectional
-representations from unlabeled text by jointly conditioning on both left and right context in all layers. As a result,
-the pre-trained BERT model can be fine-tuned with just one additional output layer to create state-of-the-art models
-for a wide range of tasks, such as question answering and language inference, without substantial task-specific
-architecture modifications.*
+> [!TIP]
+> Click on the BERT models in the right sidebar for more examples of how to apply BERT to different language tasks.
 
-*BERT is conceptually simple and empirically powerful. It obtains new state-of-the-art results on eleven natural
-language processing tasks, including pushing the GLUE score to 80.5% (7.7% point absolute improvement), MultiNLI
-accuracy to 86.7% (4.6% absolute improvement), SQuAD v1.1 question answering Test F1 to 93.2 (1.5 point absolute
-improvement) and SQuAD v2.0 Test F1 to 83.1 (5.1 point absolute improvement).*
+The example below demonstrates how to predict the `[MASK]` token with [`Pipeline`], [`AutoModel`], and from the command line.
 
-This model was contributed by [thomwolf](https://huggingface.co/thomwolf). The original code can be found [here](https://github.com/google-research/bert).
+<hfoptions id="usage">
+<hfoption id="Pipeline">
 
-## Usage tips
+```py
+import torch
+from transformers import pipeline
 
-- BERT is a model with absolute position embeddings so it's usually advised to pad the inputs on the right rather than
-  the left.
-- BERT was trained with the masked language modeling (MLM) and next sentence prediction (NSP) objectives. It is
-  efficient at predicting masked tokens and at NLU in general, but is not optimal for text generation.
-- Corrupts the inputs by using random masking, more precisely, during pretraining, a given percentage of tokens (usually 15%) is masked by:
-
-    * a special mask token with probability 0.8
-    * a random token different from the one masked with probability 0.1
-    * the same token with probability 0.1
-    
-- The model must predict the original sentence, but has a second objective: inputs are two sentences A and B (with a separation token in between). With probability 50%, the sentences are consecutive in the corpus, in the remaining 50% they are not related. The model has to predict if the sentences are consecutive or not.
-
-### Using Scaled Dot Product Attention (SDPA)
-
-PyTorch includes a native scaled dot-product attention (SDPA) operator as part of `torch.nn.functional`. This function 
-encompasses several implementations that can be applied depending on the inputs and the hardware in use. See the 
-[official documentation](https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html) 
-or the [GPU Inference](https://huggingface.co/docs/transformers/main/en/perf_infer_gpu_one#pytorch-scaled-dot-product-attention)
-page for more information.
-
-SDPA is used by default for `torch>=2.1.1` when an implementation is available, but you may also set 
-`attn_implementation="sdpa"` in `from_pretrained()` to explicitly request SDPA to be used.
-
-```
-from transformers import BertModel
-
-model = BertModel.from_pretrained("bert-base-uncased", torch_dtype=torch.float16, attn_implementation="sdpa")
-...
+pipeline = pipeline(
+    task="fill-mask",
+    model="google-bert/bert-base-uncased",
+    torch_dtype=torch.float16,
+    device=0
+)
+pipeline("Plants create [MASK] through a process known as photosynthesis.")
 ```
 
-For the best speedups, we recommend loading the model in half-precision (e.g. `torch.float16` or `torch.bfloat16`).
+</hfoption>
+<hfoption id="AutoModel">
 
-On a local benchmark (A100-80GB, CPUx12, RAM 96.6GB, PyTorch 2.2.0, OS Ubuntu 22.04) with `float16`, we saw the 
-following speedups during training and inference.
+```py
+import torch
+from transformers import AutoModelForMaskedLM, AutoTokenizer
 
-#### Training
+tokenizer = AutoTokenizer.from_pretrained(
+    "google-bert/bert-base-uncased",
+)
+model = AutoModelForMaskedLM.from_pretrained(
+    "google-bert/bert-base-uncased",
+    torch_dtype=torch.float16,
+    device_map="auto",
+    attn_implementation="sdpa"
+)
+inputs = tokenizer("Plants create [MASK] through a process known as photosynthesis.", return_tensors="pt").to("cuda")
 
-|batch_size|seq_len|Time per batch (eager - s)|Time per batch (sdpa - s)|Speedup (%)|Eager peak mem (MB)|sdpa peak mem (MB)|Mem saving (%)|
-|----------|-------|--------------------------|-------------------------|-----------|-------------------|------------------|--------------|
-|4         |256    |0.023                     |0.017                    |35.472     |939.213            |764.834           |22.800        |
-|4         |512    |0.023                     |0.018                    |23.687     |1970.447           |1227.162          |60.569        |
-|8         |256    |0.023                     |0.018                    |23.491     |1594.295           |1226.114          |30.028        |
-|8         |512    |0.035                     |0.025                    |43.058     |3629.401           |2134.262          |70.054        |
-|16        |256    |0.030                     |0.024                    |25.583     |2874.426           |2134.262          |34.680        |
-|16        |512    |0.064                     |0.044                    |46.223     |6964.659           |3961.013          |75.830        |
+with torch.no_grad():
+    outputs = model(**inputs)
+    predictions = outputs.logits
 
-#### Inference
+masked_index = torch.where(inputs['input_ids'] == tokenizer.mask_token_id)[1]
+predicted_token_id = predictions[0, masked_index].argmax(dim=-1)
+predicted_token = tokenizer.decode(predicted_token_id)
 
-|batch_size|seq_len|Per token latency eager (ms)|Per token latency SDPA (ms)|Speedup (%)|Mem eager (MB)|Mem BT (MB)|Mem saved (%)|
-|----------|-------|----------------------------|---------------------------|-----------|--------------|-----------|-------------|
-|1         |128    |5.736                       |4.987                      |15.022     |282.661       |282.924    |-0.093       |
-|1         |256    |5.689                       |4.945                      |15.055     |298.686       |298.948    |-0.088       |
-|2         |128    |6.154                       |4.982                      |23.521     |314.523       |314.785    |-0.083       |
-|2         |256    |6.201                       |4.949                      |25.303     |347.546       |347.033    |0.148        |
-|4         |128    |6.049                       |4.987                      |21.305     |378.895       |379.301    |-0.107       |
-|4         |256    |6.285                       |5.364                      |17.166     |443.209       |444.382    |-0.264       |
+print(f"The predicted token is: {predicted_token}")
+```
 
+</hfoption>
+<hfoption id="transformers-cli">
 
+```bash
+echo -e "Plants create [MASK] through a process known as photosynthesis." | transformers-cli run --task fill-mask --model google-bert/bert-base-uncased --device 0
+```
 
-## Resources
+</hfoption>
+</hfoptions>
 
-A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with BERT. If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.
+## Notes
 
-<PipelineTag pipeline="text-classification"/>
-
-- A blog post on [BERT Text Classification in a different language](https://www.philschmid.de/bert-text-classification-in-a-different-language).
-- A notebook for [Finetuning BERT (and friends) for multi-label text classification](https://colab.research.google.com/github/NielsRogge/Transformers-Tutorials/blob/master/BERT/Fine_tuning_BERT_(and_friends)_for_multi_label_text_classification.ipynb).
-- A notebook on how to [Finetune BERT for multi-label classification using PyTorch](https://colab.research.google.com/github/abhimishra91/transformers-tutorials/blob/master/transformers_multi_label_classification.ipynb). 🌎
-- A notebook on how to [warm-start an EncoderDecoder model with BERT for summarization](https://colab.research.google.com/github/patrickvonplaten/notebooks/blob/master/BERT2BERT_for_CNN_Dailymail.ipynb).
-- [`BertForSequenceClassification`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/text-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification.ipynb).
-- [`TFBertForSequenceClassification`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/text-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification-tf.ipynb).
-- [`FlaxBertForSequenceClassification`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/flax/text-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/text_classification_flax.ipynb).
-- [Text classification task guide](../tasks/sequence_classification)
-
-<PipelineTag pipeline="token-classification"/>
-
-- A blog post on how to use [Hugging Face Transformers with Keras: Fine-tune a non-English BERT for Named Entity Recognition](https://www.philschmid.de/huggingface-transformers-keras-tf).
-- A notebook for [Finetuning BERT for named-entity recognition](https://colab.research.google.com/github/NielsRogge/Transformers-Tutorials/blob/master/BERT/Custom_Named_Entity_Recognition_with_BERT_only_first_wordpiece.ipynb) using only the first wordpiece of each word in the word label during tokenization. To propagate the label of the word to all wordpieces, see this [version](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/BERT/Custom_Named_Entity_Recognition_with_BERT.ipynb) of the notebook instead.
-- [`BertForTokenClassification`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/token-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification.ipynb).
-- [`TFBertForTokenClassification`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/token-classification) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/token_classification-tf.ipynb).
-- [`FlaxBertForTokenClassification`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/flax/token-classification).
-- [Token classification](https://huggingface.co/course/chapter7/2?fw=pt) chapter of the 🤗 Hugging Face Course.
-- [Token classification task guide](../tasks/token_classification)
-
-<PipelineTag pipeline="fill-mask"/>
-
-- [`BertForMaskedLM`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/language-modeling#robertabertdistilbert-and-masked-language-modeling) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling.ipynb).
-- [`TFBertForMaskedLM`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/language-modeling#run_mlmpy) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/language_modeling-tf.ipynb).
-- [`FlaxBertForMaskedLM`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/flax/language-modeling#masked-language-modeling) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/masked_language_modeling_flax.ipynb).
-- [Masked language modeling](https://huggingface.co/course/chapter7/3?fw=pt) chapter of the 🤗 Hugging Face Course.
-- [Masked language modeling task guide](../tasks/masked_language_modeling)
-
-<PipelineTag pipeline="question-answering"/>
-
-- [`BertForQuestionAnswering`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/question-answering) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering.ipynb).
-- [`TFBertForQuestionAnswering`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/question-answering) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering-tf.ipynb).
-- [`FlaxBertForQuestionAnswering`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/flax/question-answering).
-- [Question answering](https://huggingface.co/course/chapter7/7?fw=pt) chapter of the 🤗 Hugging Face Course.
-- [Question answering task guide](../tasks/question_answering)
-
-**Multiple choice**
-- [`BertForMultipleChoice`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/multiple-choice) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/multiple_choice.ipynb).
-- [`TFBertForMultipleChoice`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/tensorflow/multiple-choice) and [notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/multiple_choice-tf.ipynb).
-- [Multiple choice task guide](../tasks/multiple_choice)
-
-⚡️ **Inference**
-- A blog post on how to [Accelerate BERT inference with Hugging Face Transformers and AWS Inferentia](https://huggingface.co/blog/bert-inferentia-sagemaker).
-- A blog post on how to [Accelerate BERT inference with DeepSpeed-Inference on GPUs](https://www.philschmid.de/bert-deepspeed-inference).
-
-⚙️ **Pretraining**
-- A blog post on [Pre-Training BERT with Hugging Face Transformers and Habana Gaudi](https://www.philschmid.de/pre-training-bert-habana).
-
-🚀 **Deploy**
-- A blog post on how to [Convert Transformers to ONNX with Hugging Face Optimum](https://www.philschmid.de/convert-transformers-to-onnx).
-- A blog post on how to [Setup Deep Learning environment for Hugging Face Transformers with Habana Gaudi on AWS](https://www.philschmid.de/getting-started-habana-gaudi#conclusion).
-- A blog post on [Autoscaling BERT with Hugging Face Transformers, Amazon SageMaker and Terraform module](https://www.philschmid.de/terraform-huggingface-amazon-sagemaker-advanced).
-- A blog post on [Serverless BERT with HuggingFace, AWS Lambda, and Docker](https://www.philschmid.de/serverless-bert-with-huggingface-aws-lambda-docker).
-- A blog post on [Hugging Face Transformers BERT fine-tuning using Amazon SageMaker and Training Compiler](https://www.philschmid.de/huggingface-amazon-sagemaker-training-compiler).
-- A blog post on [Task-specific knowledge distillation for BERT using Transformers & Amazon SageMaker](https://www.philschmid.de/knowledge-distillation-bert-transformers).
+- Inputs should be padded on the right because BERT uses absolute position embeddings.
 
 ## BertConfig
 
@@ -181,35 +107,10 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
     - create_token_type_ids_from_sequences
     - save_vocabulary
 
-<frameworkcontent>
-<pt>
-
 ## BertTokenizerFast
 
 [[autodoc]] BertTokenizerFast
 
-</pt>
-<tf>
-
-## TFBertTokenizer
-
-[[autodoc]] TFBertTokenizer
-
-</tf>
-</frameworkcontent>
-
-## Bert specific outputs
-
-[[autodoc]] models.bert.modeling_bert.BertForPreTrainingOutput
-
-[[autodoc]] models.bert.modeling_tf_bert.TFBertForPreTrainingOutput
-
-[[autodoc]] models.bert.modeling_flax_bert.FlaxBertForPreTrainingOutput
-
-
-<frameworkcontent>
-<pt>
-
 ## BertModel
 
 [[autodoc]] BertModel
@@ -255,8 +156,9 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
 [[autodoc]] BertForQuestionAnswering
     - forward
 
-</pt>
-<tf>
+## TFBertTokenizer
+
+[[autodoc]] TFBertTokenizer
 
 ## TFBertModel
 
@@ -303,9 +205,6 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
 [[autodoc]] TFBertForQuestionAnswering
     - call
 
-</tf>
-<jax>
-
 ## FlaxBertModel
 
 [[autodoc]] FlaxBertModel
@@ -351,7 +250,10 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
 [[autodoc]] FlaxBertForQuestionAnswering
     - __call__
 
-</jax>
-</frameworkcontent>
+## Bert specific outputs
 
+[[autodoc]] models.bert.modeling_bert.BertForPreTrainingOutput
 
+[[autodoc]] models.bert.modeling_tf_bert.TFBertForPreTrainingOutput
+
+[[autodoc]] models.bert.modeling_flax_bert.FlaxBertForPreTrainingOutput

docs/source/en/model_doc/deepseek_v3.md

@@ -0,0 +1,184 @@
+<!--Copyright 2025 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+
+⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+rendered properly in your Markdown viewer.
+
+-->
+
+# DeepSeek-V3
+
+## Overview
+
+The DeepSeek-V3 model was proposed in [DeepSeek-V3 Technical Report](https://arxiv.org/abs/2412.19437) by DeepSeek-AI Team.
+
+The abstract from the paper is the following:
+We present DeepSeek-V3, a strong Mixture-of-Experts (MoE) language model with 671B total parameters with 37B activated for each token. To achieve efficient inference and cost-effective training, DeepSeek-V3 adopts Multi-head Latent Attention (MLA) and DeepSeekMoE architectures, which were thoroughly validated in DeepSeek-V2. Furthermore, DeepSeek-V3 pioneers an auxiliary-loss-free strategy for load balancing and sets a multi-token prediction training objective for stronger performance. We pre-train DeepSeek-V3 on 14.8 trillion diverse and high-quality tokens, followed by Supervised Fine-Tuning and Reinforcement Learning stages to fully harness its capabilities. Comprehensive evaluations reveal that DeepSeek-V3 outperforms other open-source models and achieves performance comparable to leading closed-source models. Despite its excellent performance, DeepSeek-V3 requires only 2.788M H800 GPU hours for its full training. In addition, its training process is remarkably stable. Throughout the entire training process, we did not experience any irrecoverable loss spikes or perform any rollbacks. The model checkpoints are available at https://github.com/deepseek-ai/DeepSeek-V3.
+
+## Limitations and call for contribution!
+
+We are super happy to make this code community-powered, and would love to see how you can best optimize the following: 
+
+- current implementation uses the "naive" attention compution (so not really MLA)
+- current implementation loops through the experts. This should be replaced. Pointers to use `get_packed_weights` from `intetrations/tensor_parallel`. 
+- current implementation uses the eleuther formula for ROPE, using the orginal one would be more efficient! (should still follow our API)
+- static cache is not supported (this should be just a generation config issue / config shape issues)
+
+### Usage tips
+The model uses Multi-head Latent Attention (MLA) and DeepSeekMoE architectures for efficient inference and cost-effective training. It employs an auxiliary-loss-free strategy for load balancing and multi-token prediction training objective. The model can be used for various language tasks after being pre-trained on 14.8 trillion tokens and going through Supervised Fine-Tuning and Reinforcement Learning stages.
+
+You can run the model in `FP8` automatically, using 2 nodes of 8 H100 should be more than enough! 
+
+```python
+# `run_deepseek_v1.py`
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
+torch.manual_seed(30)
+
+tokenizer = AutoTokenizer.from_pretrained("deepseek-r1")
+
+chat = [
+  {"role": "user", "content": "Hello, how are you?"},
+  {"role": "assistant", "content": "I'm doing great. How can I help you today?"},
+  {"role": "user", "content": "I'd like to show off how chat templating works!"},
+]
+
+
+model = AutoModelForCausalLM.from_pretrained("deepseek-r1", device_map="auto", torch_dtype=torch.bfloat16)
+inputs = tokenizer.apply_chat_template(chat, tokenize=True, add_generation_prompt=True, return_tensors="pt").to(model.device)
+import time
+start = time.time()
+outputs = model.generate(inputs, max_new_tokens=50)
+print(tokenizer.batch_decode(outputs))
+print(time.time()-start)
+```
+This generated: 
+
+``````
+<｜Assistant｜><think>
+Okay, the user wants to demonstrate how chat templating works. Let me break down what that means. Chat templating is about structuring the conversation data, especially for models that need specific input formats. Maybe they're referring to something like how messages are formatted with roles (user, assistant, system) in APIs like OpenAI.
+
+First, I should explain what chat templating is. It's the process of formatting conversation data into a structured format that the model can understand. This usually includes roles and content. For example, user messages, assistant responses, and system messages each have their own role tags.
+
+They might want an example. Let me think of a simple conversation. The user says "Hello, how are you?" and the assistant responds "I'm doing great. How can I help you today?" Then the user follows up with wanting to show off chat templating. So the example should include the history and the new message.
+
+In some frameworks, like Hugging Face's Transformers, chat templates are applied using Jinja2 templates. The template might look something like combining system messages, then looping through user and assistant messages with appropriate tags. For instance, using {% for message in messages %} and assigning roles like <|user|>, <|assistant|>, etc.
+
+I should structure the example with the messages array, showing each role and content. Then apply a hypothetical template to convert that into a formatted string the model uses. Also, mention that different models have different templating requirements, like using special tokens or varying role labels.
+
+Wait, the user mentioned "chat templating" in the context of showing off. Maybe they want a practical example they can present. So providing a code snippet or a structured data example would be helpful. Let me outline a typical messages array and then the templated output.
+
+Also, it's important to note that proper templating ensures the model knows the conversation flow, which is crucial for generating coherent responses. Maybe include a note about why it's important, like maintaining context and role-specific processing.
+
+Let me check if there are any common mistakes or things to avoid. For example, not closing tags properly, or mismatching roles. But maybe that's too detailed unless the user asks. Focus on the positive example first.
+
+Putting it all together, the response should have an example messages array, the applied template, and the final formatted string. Maybe use angle brackets or special tokens as placeholders. Also, mention that this helps in training or fine-tuning models with structured data.
+
+I think that's a solid approach. Let me structure it step by step to make it clear.
+</think>
+
+Chat templating is a way to structure conversation data (e.g., user/assistant interactions) into a format that language models understand. This is especially important for models trained to handle multi-turn dialogues, where the input must explicitly separate roles (user, assistant, system, etc.) and messages. Let’s break this down with an example!
+
+---
+
+### **Step 1: Raw Conversation History**
+Suppose we have this conversation:
+- **User**: "Hello, how are you?"
+- **Assistant**: "I'm doing great. How can I help you today?"
+- **User**: "I'd like to show off how chat templating works!"
+
+---
+
+### **Step 2: Structured Messages**
+In frameworks like Hugging Face Transformers or OpenAI, conversations are often formatted as a list of dictionaries with `role` and `content`:
+```python
+messages = [
+    {"role": "user", "content": "Hello, how are you?"},
+    {"role": "assistant", "content": "I'm doing great. How can I help you today?"},
+    {"role": "user", "content": "I'd like to show off how chat templating works!"},
+]
+```
+
+---
+
+### **Step 3: Apply a Chat Template**
+A **chat template** converts this structured data into a single string formatted for the model. For example, using a Jinja-style template (common in Hugging Face):
+
+```jinja
+{% for message in messages %}
+    {% if message['role'] == 'user' %}
+        <|user|>{{ message['content'] }}<|end|>
+    {% elif message['role'] == 'assistant' %}
+        <|assistant|>{{ message['content'] }}<|end|>
+    {% endif %}
+{% endfor %}
+<|assistant|>
+```
+
+---
+
+### **Step 4: Final Templated Output**
+Applying the template to our `messages` list would produce:
+```text
+<|user|>Hello, how are you?<|end|>
+<|assistant|>I'm doing great. How can I help you today?<|end|>
+<|user|>I'd like to show off how chat templating works!<|end|>
+<|assistant|>
+```
+
+This tells the model:  
+1. The conversation history (user/assistant turns).  
+2. The model’s turn to generate a response (`<|assistant|>` at the end).  
+
+---
+
+### **Key Notes**:
+- **Role Separation**: Tags like `<|user|>` and `<|assistant|>` help the model distinguish speakers.
+- **Special Tokens**: Models often use unique tokens (e.g., `<|end|>`) to mark message boundaries.
+- **Flexibility**: Templates vary by model (e.g., OpenAI uses `{"role": "user", "content": "..."}` instead of tags).
+
+---
+
+### **Why This Matters**:
+- **Consistency**: Ensures the model understands dialogue structure.
+- **Context Preservation**: Maintains the flow of multi-turn conversations.
+- **Alignment**: Matches the format the model was trained on for better performance.
+
+Want to dive deeper or see a specific framework’s implementation (e.g., OpenAI, Llama, Mistral)? Let me know! 😊<｜end▁of▁sentence｜>
+``````
+
+Use the following to run it
+```bash
+torchrun --nproc_per_node=8 --nnodes=2 --node_rank=0|1 --rdzv-id an_id --rdzv-backend c10d --rdzv-endpoint master_addr:master_port run_deepseek_r1.py
+```
+
+If you have: 
+```bash
+[rank0]: ncclInternalError: Internal check failed.
+[rank0]: Last error:
+[rank0]: Bootstrap : no socket interface found
+```
+error, it means NCCL was probably not loaded. 
+
+
+## DeepseekV3Config
+
+[[autodoc]] DeepseekV3Config
+
+## DeepseekV3Model
+
+[[autodoc]] DeepseekV3Model
+    - forward
+
+## DeepseekV3ForCausalLM
+
+[[autodoc]] DeepseekV3ForCausalLM
+    - forward

docs/source/en/model_doc/gemma3.md

@@ -15,74 +15,63 @@ rendered properly in your Markdown viewer.
 
 -->
 
-# Gemma3
+<div style="float: right;">
+    <div class="flex flex-wrap space-x-1">
+        <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
+        <img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+    </div>
+</div>
 
-## Overview
+# Gemma 3
 
-The Gemma 3 model was proposed in the [Gemma 3 Techncial Report](https://goo.gle/Gemma3Report) by Google. It is a vision-language model composed by a [SigLIP](siglip) vision encoder and a [Gemma 2](gemma_2) language decoder, linked by a multimodal linear projection. It cuts an image into a fixed number of tokens, in the same way as SigLIP, as long as the image does not exceed certain aspect ratio. For images that exceed the given aspect ratio, it crops the image into multiple smaller patches and concatenates them with the base image embedding. One particularity is that the model uses bidirectional attention on all the image tokens. In addition, the model interleaves sliding window local attention with full causal attention in the language backbone, where each sixth layer is a full causal attention layer.
+[Gemma 3](https://goo.gle/Gemma3Report) is a multimodal model with pretrained and instruction-tuned variants, available in 1B, 13B, and 27B parameters. The architecture is mostly the same as the previous Gemma versions. The key differences are alternating 5 local sliding window self-attention layers for every global self-attention layer, support for a longer context length of 128K tokens, and a [SigLip](./siglip) encoder that can "pan & scan" high-resolution images to prevent information from disappearing in high resolution images or images with non-square aspect ratios.
 
-This model was contributed by [Ryan Mullins](https://huggingface.co/RyanMullins), [Raushan Turganbay](https://huggingface.co/RaushanTurganbay) [Arthur Zucker](https://huggingface.co/ArthurZ), and [Pedro Cuenca](https://huggingface.co/pcuenq).
+The instruction-tuned variant was post-trained with knowledge distillation and reinforcement learning.
 
+You can find all the original Gemma 3 checkpoints under the [Gemma 3](https://huggingface.co/collections/meta-llama/llama-2-family-661da1f90a9d678b6f55773b) release.
 
-## Usage tips
+> [!TIP]
+> Click on the Gemma 3 models in the right sidebar for more examples of how to apply Gemma to different vision and language tasks.
 
+The example below demonstrates how to generate text based on an image with [`Pipeline`] or the [`AutoModel`] class.
 
-- For image+text and image-only inputs use `Gemma3ForConditionalGeneration`.
-- For text-only inputs use `Gemma3ForCausalLM` for generation to avoid loading the vision tower.
-- Each sample can contain multiple images, and the number of images can vary between samples. However, make sure to pass correctly batched images to the processor, where each batch is a list of one or more images.
-- The text passed to the processor should have a `<start_of_image>` token wherever an image should be inserted.
-- The processor has its own `apply_chat_template` method to convert chat messages to model inputs. See the examples below for more details on how to use it.
+<hfoptions id="usage">
+<hfoption id="Pipeline">
 
+```py
+import torch
+from transformers import pipeline
 
-### Image cropping for high resolution images
-
-The model supports cropping images into smaller patches when the image aspect ratio exceeds a certain value. By default the images are not cropped and only the base image is forwarded to the model. Users can set `do_pan_and_scan=True` to obtain several crops per image along with the base image to improve the quality in DocVQA or similar tasks requiring higher resolution images.
-
-Pan and scan is an inference time optimization to handle images with skewed aspect ratios. When enabled, it improves performance on tasks related to document understanding, infographics, OCR, etc.
-
-```python
-
-processor = AutoProcessor.from_pretrained("google/gemma-3-4b-it", padding_side="left")
-
-url = "https://media.istockphoto.com/id/1192867753/photo/cow-in-berchida-beach-siniscola.jpg?s=612x612&w=0&k=20&c=v0hjjniwsMNfJSuKWZuIn8pssmD5h5bSN1peBd1CmH4="
-messages = [
-    {
-        "role": "system",
-        "content": [
-            {"type": "text", "text": "You are a helpful assistant."}
-        ]
-    },
-    {
-        "role": "user", "content": [
-            {"type": "image", "url": url},
-            {"type": "text", "text": "What is shown in this image?"},
-        ]
-    },
-]
-inputs = processor.apply_chat_template(
-    messages,
-    tokenize=True,
-    return_dict=True,
-    return_tensors="pt",
-    add_generation_prompt=True,
-    do_pan_and_scan=True,
-).to(model.device)
-
+pipeline = pipeline(
+    task="image-text-to-text",
+    model="google/gemma-3-4b-pt",
+    device=0,
+    torch_dtype=torch.bfloat16
+)
+pipeline(
+    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg",
+    text="<start_of_image> What is shown in this image?"
+)
 ```
 
+</hfoption>
+<hfoption id="AutoModel">
 
-## Usage Example
-
-### Single-image Inference
-
-```python
+```py
+import torch
 from transformers import AutoProcessor, Gemma3ForConditionalGeneration
 
-model_id = "google/gemma-3-4b-it"
-model = Gemma3ForConditionalGeneration.from_pretrained(model_id, device_map="auto")
-processor = AutoProcessor.from_pretrained(model_id, padding_side="left")
+model = Gemma3ForConditionalGeneration.from_pretrained(
+    "google/gemma-3-4b-it",
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    attn_implementation="sdpa"
+)
+processor = AutoProcessor.from_pretrained(
+    "google/gemma-3-4b-it",
+    padding_side="left"
+)
 
-url = "https://media.istockphoto.com/id/1192867753/photo/cow-in-berchida-beach-siniscola.jpg?s=612x612&w=0&k=20&c=v0hjjniwsMNfJSuKWZuIn8pssmD5h5bSN1peBd1CmH4="
 messages = [
     {
         "role": "system",
@@ -92,7 +81,7 @@ messages = [
     },
     {
         "role": "user", "content": [
-            {"type": "image", "url": url},
+            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"},
             {"type": "text", "text": "What is shown in this image?"},
         ]
     },
@@ -103,21 +92,43 @@ inputs = processor.apply_chat_template(
     return_dict=True,
     return_tensors="pt",
     add_generation_prompt=True,
-).to(model.device)
+).to("cuda")
 
-output = model.generate(**inputs, max_new_tokens=50)
-print(processor.decode(output[0], skip_special_tokens=True)[inputs.input_ids.shape[1]: ])
+output = model.generate(**inputs, max_new_tokens=50, cache_implementation="static")
+print(processor.decode(output[0], skip_special_tokens=True))
 ```
 
-### Multi-image Inference
+</hfoption>
+<hfoption id="transformers-cli">
 
-```python
-model_id = "google/gemma-3-4b-it"
-model = Gemma3ForConditionalGeneration.from_pretrained(model_id, device_map="auto")
-processor = AutoProcessor.from_pretrained(model_id, padding_side="left")
+```bash
+echo -e "Plants create energy through a process known as" | transformers-cli run --task text-generation --model google/gemma-3-1b-pt --device 0
+```
+
+</hfoption>
+</hfoptions>
+
+Quantization reduces the memory burden of large models by representing the weights in a lower precision. Refer to the [Quantization](../quantization/overview) overview for more available quantization backends.
+
+The example below uses [torchao](../quantization/torchao) to only quantize the weights to int4.
+
+```py
+# pip install torchao
+import torch
+from transformers import TorchAoConfig, Gemma3ForConditionalGeneration, AutoProcessor
+
+quantization_config = TorchAoConfig("int4_weight_only", group_size=128)
+model = Gemma3ForConditionalGeneration.from_pretrained(
+    "google/gemma-3-27b-it",
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    quantization_config=quantization_config
+)
+processor = AutoProcessor.from_pretrained(
+    "google/gemma-3-27b-it",
+    padding_side="left"
+)
 
-url_cow = "https://media.istockphoto.com/id/1192867753/photo/cow-in-berchida-beach-siniscola.jpg?s=612x612&w=0&k=20&c=v0hjjniwsMNfJSuKWZuIn8pssmD5h5bSN1peBd1CmH4="
-url_stop = "https://www.ilankelman.org/stopsigns/australia.jpg"
 messages = [
     {
         "role": "system",
@@ -127,9 +138,8 @@ messages = [
     },
     {
         "role": "user", "content": [
-            {"type": "image", "url": url_cow},
-            {"type": "image", "url": url_stop},
-            {"type": "text", "text": "Are these two images identical?"},
+            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"},
+            {"type": "text", "text": "What is shown in this image?"},
         ]
     },
 ]
@@ -139,33 +149,85 @@ inputs = processor.apply_chat_template(
     return_dict=True,
     return_tensors="pt",
     add_generation_prompt=True,
-).to(model.device)
-
-output = model.generate(**inputs, max_new_tokens=50)
-print(processor.decode(output[0], skip_special_tokens=True)[inputs.input_ids.shape[1]: ])
+).to("cuda")
 
+output = model.generate(**inputs, max_new_tokens=50, cache_implementation="static")
+print(processor.decode(output[0], skip_special_tokens=True))
 ```
 
-### Text-only inference
+Use the [AttentionMaskVisualizer](https://github.com/huggingface/transformers/blob/beb9b5b02246b9b7ee81ddf938f93f44cfeaad19/src/transformers/utils/attention_visualizer.py#L139) to better understand what tokens the model can and cannot attend to.
 
-You can use the VLMs for text-only generation by omitting images in your input. However, you can also load the models in text-only mode as shown below. This will skip loading the vision tower and will save resources when you just need the LLM capabilities.
-```python
-from transformers import AutoTokenizer, Gemma3ForCausalLM
-
-model_id = "google/gemma-3-1b-it"
-
-tokenizer = AutoTokenizer.from_pretrained(model_id)
-model = Gemma3ForCausalLM.from_pretrained(model_id, device_map="auto")
-
-input_ids = tokenizer("Write me a poem about Machine Learning.", return_tensors="pt").to(model.device)
-
-outputs = model.generate(**input_ids, max_new_tokens=100)
-text = tokenizer.batch_decode(outputs, skip_special_tokens=True)
-
-print(text)
+```py
+from transformers.utils.attention_visualizer import AttentionMaskVisualizer
 
+visualizer = AttentionMaskVisualizer("google/gemma-3-4b-it")
+visualizer("<img>What is shown in this image?")
 ```
 
+<div class="flex justify-center">
+    <img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/gemma-3-attn-mask.png"/>
+</div>
+
+## Notes
+
+- Use [`Gemma3ForConditionalGeneration`] for image-and-text and image-only inputs.
+- Gemma 3 supports multiple input images, but make sure the images are correctly batched before passing them to the processor. Each batch should be a list of one or more images.
+
+    ```py
+    url_cow = "https://media.istockphoto.com/id/1192867753/photo/cow-in-berchida-beach-siniscola.jpg?s=612x612&w=0&k=20&c=v0hjjniwsMNfJSuKWZuIn8pssmD5h5bSN1peBd1CmH4="
+    url_cat = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"
+
+    messages =[
+        {
+            "role": "system",
+            "content": [
+                {"type": "text", "text": "You are a helpful assistant."}
+            ]
+        },
+        {
+            "role": "user",
+            "content": [
+                {"type": "image", "url": url_cow},
+                {"type": "image", "url": url_cat},
+                {"type": "text", "text": "Which image is cuter?"},
+            ]
+        },
+    ]
+    ```
+- Text passed to the processor should have a `<start_of_image>` token wherever an image should be inserted.
+- The processor has its own [`~ProcessorMixin.apply_chat_template`] method to convert chat messages to model inputs.
+- By default, images aren't cropped and only the base image is forwarded to the model. In high resolution images or images with non-square aspect ratios, artifacts can result because the vision encoder uses a fixed resolution of 896x896. To prevent these artifacts and improve performance during inference, set `do_pan_and_scan=True` to crop the image into multiple smaller patches and concatenate them with the base image embedding. You can disable pan and scan for faster inference.
+
+    ```diff
+    inputs = processor.apply_chat_template(
+        messages,
+        tokenize=True,
+        return_dict=True,
+        return_tensors="pt",
+        add_generation_prompt=True,
+    +   do_pan_and_scan=True,
+        ).to("cuda")
+    ```
+- For Gemma-3 1B checkpoint trained in text-only mode, use [`AutoModelForCausalLM`] instead.
+
+    ```py
+    import torch
+    from transformers import AutoModelForCausalLM, AutoTokenizer
+
+    tokenizer = AutoTokenizer.from_pretrained(
+        "google/gemma-3-1b-pt",
+    )
+    model = AutoModelForCausalLM.from_pretrained(
+        "google/gemma-3-1b-pt",
+        torch_dtype=torch.bfloat16,
+        device_map="auto",
+        attn_implementation="sdpa"
+    )
+    input_ids = tokenizer("Plants create energy through a process known as", return_tensors="pt").to("cuda")
+
+    output = model.generate(**input_ids, cache_implementation="static")
+    print(tokenizer.decode(output[0], skip_special_tokens=True))
+    ```
 
 ## Gemma3ImageProcessor
 

docs/source/en/model_doc/llama.md

@@ -14,79 +14,115 @@ rendered properly in your Markdown viewer.
 
 -->
 
-# LLaMA
-
-<div class="flex flex-wrap space-x-1">
-<img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
-<img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,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
-">
-<img alt="FlashAttention" src="https://img.shields.io/badge/%E2%9A%A1%EF%B8%8E%20FlashAttention-eae0c8?style=flat">
-<img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+<div style="float: right;">
+    <div class="flex flex-wrap space-x-1">
+        <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
+        <img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,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
+        ">
+        <img alt="FlashAttention" src="https://img.shields.io/badge/%E2%9A%A1%EF%B8%8E%20FlashAttention-eae0c8?style=flat">
+        <img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+    </div>
 </div>
 
-## Overview
+# Llama
 
-The LLaMA model was proposed in [LLaMA: Open and Efficient Foundation Language Models](https://arxiv.org/abs/2302.13971) by Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aurelien Rodriguez, Armand Joulin, Edouard Grave, Guillaume Lample. It is a collection of foundation language models ranging from 7B to 65B parameters.
+[Llama](https://huggingface.co/papers/2302.13971) is a family of large language models ranging from 7B to 65B parameters. These models are focused on efficient inference (important for serving language models) by training a smaller model on more tokens rather than training a larger model on fewer tokens. The Llama model is based on the GPT architecture, but it uses pre-normalization to improve training stability, replaces ReLU with SwiGLU to improve performance, and replaces absolute positional embeddings with rotary positional embeddings (RoPE) to better handle longer sequence lengths.
 
-The abstract from the paper is the following:
+You can find all the original Llama checkpoints under the [Huggy Llama](https://huggingface.co/huggyllama) organization.
 
-*We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. We train our models on trillions of tokens, and show that it is possible to train state-of-the-art models using publicly available datasets exclusively, without resorting to proprietary and inaccessible datasets. In particular, LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B. We release all our models to the research community. *
+> [!TIP]
+> Click on the Llama models in the right sidebar for more examples of how to apply Llama to different language tasks.
 
-This model was contributed by [zphang](https://huggingface.co/zphang) with contributions from [BlackSamorez](https://huggingface.co/BlackSamorez). The code of the implementation in Hugging Face is based on GPT-NeoX [here](https://github.com/EleutherAI/gpt-neox). The original code of the authors can be found [here](https://github.com/facebookresearch/llama).
+The example below demonstrates how to generate text with [`Pipeline`] or the [`AutoModel`], and from the command line.
 
-## Usage tips
+<hfoptions id="usage">
+<hfoption id="Pipeline">
 
-- Weights for the LLaMA models can be obtained from by filling out [this form](https://docs.google.com/forms/d/e/1FAIpQLSfqNECQnMkycAp2jP4Z9TFX0cGR4uf7b_fBxjY_OjhJILlKGA/viewform?usp=send_form)
-- After downloading the weights, they will need to be converted to the Hugging Face Transformers format using the [conversion script](https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/convert_llama_weights_to_hf.py). The script can be called with the following (example) command:
+```py
+import torch
+from transformers import pipeline
+
+pipeline = pipeline(
+    task="text-generation",
+    model="huggyllama/llama-7b",
+    torch_dtype=torch.float16,
+    device=0
+)
+pipeline("Plants create energy through a process known as")
+```
+
+</hfoption>
+<hfoption id="AutoModel">
+
+```py
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+tokenizer = AutoTokenizer.from_pretrained(
+    "huggyllama/llama-7b",
+)
+model = AutoModelForCausalLM.from_pretrained(
+    "huggyllama/llama-7b",
+    torch_dtype=torch.float16,
+    device_map="auto",
+    attn_implementation="sdpa"
+)
+input_ids = tokenizer("Plants create energy through a process known as", return_tensors="pt").to("cuda")
+
+output = model.generate(**input_ids, cache_implementation="static")
+print(tokenizer.decode(output[0], skip_special_tokens=True))
+```
+
+</hfoption>
+<hfoption id="transformers-cli">
 
 ```bash
-python src/transformers/models/llama/convert_llama_weights_to_hf.py \
-    --input_dir /path/to/downloaded/llama/weights --model_size 7B --output_dir /output/path
+echo -e "Plants create energy through a process known as" | transformers-cli run --task text-generation --model huggyllama/llama-7b --device 0
 ```
 
-- After conversion, the model and tokenizer can be loaded via:
+</hfoption>
+</hfoptions>
 
-```python
-from transformers import LlamaForCausalLM, LlamaTokenizer
+Quantization reduces the memory burden of large models by representing the weights in a lower precision. Refer to the [Quantization](../quantization/overview) overview for more available quantization backends.
 
-tokenizer = LlamaTokenizer.from_pretrained("/output/path")
-model = LlamaForCausalLM.from_pretrained("/output/path")
+The example below uses [torchao](../quantization/torchao) to only quantize the weights to int4.
+
+```py
+# pip install torchao
+import torch
+from transformers import TorchAoConfig, AutoModelForCausalLM, AutoTokenizer
+
+quantization_config = TorchAoConfig("int4_weight_only", group_size=128)
+model = AutoModelForCausalLM.from_pretrained(
+    "huggyllama/llama-30b",
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    quantization_config=quantization_config
+)
+
+tokenizer = AutoTokenizer.from_pretrained("huggyllama/llama-30b")
+input_ids = tokenizer("Plants create energy through a process known as", return_tensors="pt").to("cuda")
+
+output = model.generate(**input_ids, cache_implementation="static")
+print(tokenizer.decode(output[0], skip_special_tokens=True))
 ```
 
-Note that executing the script requires enough CPU RAM to host the whole model in float16 precision (even if the biggest versions
-come in several checkpoints they each contain a part of each weight of the model, so we need to load them all in RAM). For the 65B model, it's thus 130GB of RAM needed.
+Use the [AttentionMaskVisualizer](https://github.com/huggingface/transformers/blob/beb9b5b02246b9b7ee81ddf938f93f44cfeaad19/src/transformers/utils/attention_visualizer.py#L139) to better understand what tokens the model can and cannot attend to.
 
-- The LLaMA tokenizer is a BPE model based on [sentencepiece](https://github.com/google/sentencepiece). One quirk of sentencepiece is that when decoding a sequence, if the first token is the start of the word (e.g. "Banana"), the tokenizer does not prepend the prefix space to the string.
+```py
+from transformers.utils.attention_visualizer import AttentionMaskVisualizer
 
-This model was contributed by [zphang](https://huggingface.co/zphang) with contributions from [BlackSamorez](https://huggingface.co/BlackSamorez). The code of the implementation in Hugging Face is based on GPT-NeoX [here](https://github.com/EleutherAI/gpt-neox). The original code of the authors can be found [here](https://github.com/facebookresearch/llama). The Flax version of the implementation was contributed by [afmck](https://huggingface.co/afmck) with the code in the implementation based on Hugging Face's Flax GPT-Neo.
+visualizer = AttentionMaskVisualizer("huggyllama/llama-7b")
+visualizer("Plants create energy through a process known as")
+```
 
+<div class="flex justify-center">
+    <img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/llama-attn-mask.png"/>
+</div>
 
-Based on the original LLaMA model, Meta AI has released some follow-up works:
+## Notes
 
-- **Llama2**: Llama2 is an improved version of Llama with some architectural tweaks (Grouped Query Attention), and is pre-trained on 2Trillion tokens. Refer to the documentation of Llama2 which can be found [here](llama2).
-
-## Resources
-
-A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with LLaMA. If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.
-
-<PipelineTag pipeline="text-classification"/>
-
-- A [notebook](https://colab.research.google.com/github/bigscience-workshop/petals/blob/main/examples/prompt-tuning-sst2.ipynb#scrollTo=f04ba4d2) on how to use prompt tuning to adapt the LLaMA model for text classification task. 🌎
-
-<PipelineTag pipeline="question-answering"/>
-
-- [StackLLaMA: A hands-on guide to train LLaMA with RLHF](https://huggingface.co/blog/stackllama#stackllama-a-hands-on-guide-to-train-llama-with-rlhf), a blog post about how to train LLaMA to answer questions on [Stack Exchange](https://stackexchange.com/) with RLHF.
-
-⚗️ Optimization
-- A [notebook](https://colab.research.google.com/drive/1SQUXq1AMZPSLD4mk3A3swUIc6Y2dclme?usp=sharing) on how to fine-tune LLaMA model using xturing library on GPU which has limited memory. 🌎 
-
-⚡️ Inference
-- A [notebook](https://colab.research.google.com/github/DominguesM/alpaca-lora-ptbr-7b/blob/main/notebooks/02%20-%20Evaluate.ipynb) on how to run the LLaMA Model using PeftModel from the 🤗 PEFT library. 🌎 
-- A [notebook](https://colab.research.google.com/drive/1l2GiSSPbajVyp2Nk3CFT4t3uH6-5TiBe?usp=sharing) on how to load a PEFT adapter LLaMA model with LangChain. 🌎
-
-🚀 Deploy
-- A [notebook](https://colab.research.google.com/github/lxe/simple-llama-finetuner/blob/master/Simple_LLaMA_FineTuner.ipynb#scrollTo=3PM_DilAZD8T) on how to fine-tune LLaMA model using LoRA method via the 🤗 PEFT library with intuitive UI. 🌎 
-- A [notebook](https://github.com/aws/amazon-sagemaker-examples/blob/main/introduction_to_amazon_algorithms/jumpstart-foundation-models/text-generation-open-llama.ipynb) on how to deploy Open-LLaMA model for text generation on Amazon SageMaker. 🌎 
+- The tokenizer is a byte-pair encoding model based on [SentencePiece](https://github.com/google/sentencepiece). During decoding, if the first token is the start of the word (for example, "Banana"), the tokenizer doesn't prepend the prefix space to the string.
 
 ## LlamaConfig
 

docs/source/en/model_doc/llama2.md

@@ -14,97 +14,129 @@ rendered properly in your Markdown viewer.
 
 -->
 
-# Llama2
-
-<div class="flex flex-wrap space-x-1">
-<img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
-<img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,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
-">
+<div style="float: right;">
+    <div class="flex flex-wrap space-x-1">
+        <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
+        <img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,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
+        ">
+    </div>
 </div>
 
-## Overview
+# Llama 2
 
-The Llama2 model was proposed in [LLaMA: Open Foundation and Fine-Tuned Chat Models](https://ai.meta.com/research/publications/llama-2-open-foundation-and-fine-tuned-chat-models/) by Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, Dan Bikel, Lukas Blecher, Cristian Canton Ferrer, Moya Chen, Guillem Cucurull, David Esiobu, Jude Fernandes, Jeremy Fu, Wenyin Fu, Brian Fuller, Cynthia Gao, Vedanuj Goswami, Naman Goyal, Anthony Hartshorn, Saghar Hosseini, Rui Hou, Hakan Inan, Marcin Kardas, Viktor Kerkez Madian Khabsa, Isabel Kloumann, Artem Korenev, Punit Singh Koura, Marie-Anne Lachaux, Thibaut Lavril, Jenya Lee, Diana Liskovich, Yinghai Lu, Yuning Mao, Xavier Martinet, Todor Mihaylov, Pushka rMishra, Igor Molybog, Yixin Nie, Andrew Poulton, Jeremy Reizenstein, Rashi Rungta, Kalyan Saladi, Alan Schelten, Ruan Silva, Eric Michael Smith, Ranjan Subramanian, Xiaoqing EllenTan, Binh Tang, Ross Taylor, Adina Williams, Jian Xiang Kuan, Puxin Xu, Zheng Yan, Iliyan Zarov, Yuchen Zhang, Angela Fan, Melanie Kambadur, Sharan Narang, Aurelien Rodriguez, Robert Stojnic, Sergey Edunov, Thomas Scialom. It is a collection of foundation language models ranging from 7B to 70B parameters, with checkpoints finetuned for chat application!
+[Llama 2](https://huggingface.co/papers/2307.09288) is a family of large language models, Llama 2 and Llama 2-Chat, available in 7B, 13B, and 70B parameters. The Llama 2 model mostly keeps the same architecture as [Llama](./llama), but it is pretrained on more tokens, doubles the context length, and uses grouped-query attention (GQA) in the 70B model to improve inference.
 
-The abstract from the paper is the following:
+Llama 2-Chat is trained with supervised fine-tuning (SFT), and reinforcement learning with human feedback (RLHF) - rejection sampling and proximal policy optimization (PPO) - is applied to the fine-tuned model to align the chat model with human preferences.
 
-*In this work, we develop and release Llama 2, a collection of pretrained and fine-tuned large language models (LLMs) ranging in scale from 7 billion to 70 billion parameters. Our fine-tuned LLMs, called Llama 2-Chat, are optimized for dialogue use cases. Our models outperform open-source chat models on most benchmarks we tested, and based on our human evaluations for helpfulness and safety, may be a suitable substitute for closed-source models. We provide a detailed description of our approach to fine-tuning and safety improvements of Llama 2-Chat in order to enable the community to build on our work and contribute to the responsible development of LLMs.*
+You can find all the original Llama 2 checkpoints under the [Llama 2 Family](https://huggingface.co/collections/meta-llama/llama-2-family-661da1f90a9d678b6f55773b) collection.
 
-Checkout all Llama2 model checkpoints [here](https://huggingface.co/models?search=llama2).
-This model was contributed by [Arthur Zucker](https://huggingface.co/ArthurZ) with contributions from [Lysandre Debut](https://huggingface.co/lysandre). The code of the implementation in Hugging Face is based on GPT-NeoX [here](https://github.com/EleutherAI/gpt-neox). The original code of the authors can be found [here](https://github.com/facebookresearch/llama).
+> [!TIP]
+> Click on the Llama 2 models in the right sidebar for more examples of how to apply Llama to different language tasks.
 
-## Usage tips
+The example below demonstrates how to generate text with [`Pipeline`], [`AutoModel`], and how to chat with Llama 2-Chat from the command line.
 
-<Tip warning={true}>
+<hfoptions id="usage">
+<hfoption id="Pipeline">
 
-The `Llama2` models were trained using `bfloat16`, but the original inference uses `float16`. The checkpoints uploaded on the Hub use `torch_dtype = 'float16'`, which will be
-used by the `AutoModel` API to cast the checkpoints from `torch.float32` to `torch.float16`. 
+```py
+import torch
+from transformers import pipeline
 
-The `dtype` of the online weights is mostly irrelevant unless you are using `torch_dtype="auto"` when initializing a model using `model = AutoModelForCausalLM.from_pretrained("path", torch_dtype = "auto")`. The reason is that the model will first be downloaded ( using the `dtype` of the checkpoints online), then it will be casted to the default `dtype` of `torch` (becomes `torch.float32`), and finally, if there is a `torch_dtype` provided in the config, it will be used. 
+pipeline = pipeline(
+    task="text-generation",
+    model="meta-llama/Llama-2-7b-hf",
+    torch_dtype=torch.float16,
+    device=0
+)
+pipeline("Plants create energy through a process known as")
+```
 
-Training the model in `float16` is not recommended and is known to produce `nan`; as such, the model should be trained in `bfloat16`.
+</hfoption>
+<hfoption id="AutoModel">
 
-</Tip>
+```py
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
 
-Tips:
+tokenizer = AutoTokenizer.from_pretrained(
+    "meta-llama/Llama-2-7b-hf",
+)
+model = AutoModelForCausalLM.from_pretrained(
+    "meta-llama/Llama-2-7b-hf",
+    torch_dtype=torch.float16,
+    device_map="auto",
+    attn_implementation="sdpa"
+)
+input_ids = tokenizer("Plants create energy through a process known as", return_tensors="pt").to("cuda")
 
-- Weights for the Llama2 models can be obtained by filling out [this form](https://ai.meta.com/resources/models-and-libraries/llama-downloads/)
-- The architecture is very similar to the first Llama, with the addition of Grouped Query Attention (GQA) following this [paper](https://arxiv.org/pdf/2305.13245.pdf)
-- Setting `config.pretraining_tp` to a value different than 1 will activate the more accurate but slower computation of the linear layers, which should better match the original logits.
-- The original model uses `pad_id = -1` which means that there is no padding token. We can't have the same logic, make sure to add a padding token using `tokenizer.add_special_tokens({"pad_token":"<pad>"})` and resize the token embedding accordingly. You should also set the `model.config.pad_token_id`. The `embed_tokens` layer of the model is initialized with `self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.config.padding_idx)`, which makes sure that encoding the padding token will output zeros, so passing it when initializing is recommended.
-- After filling out the form and gaining access to the model checkpoints, you should be able to use the already converted checkpoints. Otherwise, if you are converting your own model, feel free to use the [conversion script](https://github.com/huggingface/transformers/blob/main/src/transformers/models/llama/convert_llama_weights_to_hf.py). The script can be called with the following (example) command:
+output = model.generate(**input_ids, cache_implementation="static")
+print(tokenizer.decode(output[0], skip_special_tokens=True))
+```
+
+</hfoption>
+<hfoption id="transformers-cli">
 
 ```bash
-python src/transformers/models/llama/convert_llama_weights_to_hf.py \
-    --input_dir /path/to/downloaded/llama/weights --model_size 7B --output_dir /output/path
+transformers-cli chat --model_name_or_path meta-llama/Llama-2-7b-chat-hf --torch_dtype auto --attn_implementation flash_attention_2
 ```
 
-- After conversion, the model and tokenizer can be loaded via:
+</hfoption>
+</hfoptions>
 
-```python
-from transformers import LlamaForCausalLM, LlamaTokenizer
+Quantization reduces the memory burden of large models by representing the weights in a lower precision. Refer to the [Quantization](../quantization/overview) overview for more available quantization backends.
 
-tokenizer = LlamaTokenizer.from_pretrained("/output/path")
-model = LlamaForCausalLM.from_pretrained("/output/path")
+The example below uses [torchao](../quantization/torchao) to only quantize the weights to int4.
+
+```py
+# pip install torchao
+import torch
+from transformers import TorchAoConfig, AutoModelForCausalLM, AutoTokenizer
+
+quantization_config = TorchAoConfig("int4_weight_only", group_size=128)
+model = AutoModelForCausalLM.from_pretrained(
+    "meta-llama/Llama-2-13b-hf",
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    quantization_config=quantization_config
+)
+
+tokenizer = AutoTokenizer.from_pretrained("meta-llama/Llama-2-13b-hf")
+input_ids = tokenizer("Plants create energy through a process known as", return_tensors="pt").to("cuda")
+
+output = model.generate(**input_ids, cache_implementation="static")
+print(tokenizer.decode(output[0], skip_special_tokens=True))
 ```
 
-Note that executing the script requires enough CPU RAM to host the whole model in float16 precision (even if the biggest versions
-come in several checkpoints they each contain a part of each weight of the model, so we need to load them all in RAM). For the 75B model, it's thus 145GB of RAM needed.
+Use the [AttentionMaskVisualizer](https://github.com/huggingface/transformers/blob/beb9b5b02246b9b7ee81ddf938f93f44cfeaad19/src/transformers/utils/attention_visualizer.py#L139) to better understand what tokens the model can and cannot attend to.
 
-- The LLaMA tokenizer is a BPE model based on [sentencepiece](https://github.com/google/sentencepiece). One quirk of sentencepiece is that when decoding a sequence, if the first token is the start of the word (e.g. "Banana"), the tokenizer does not prepend the prefix space to the string.
+```py
+from transformers.utils.attention_visualizer import AttentionMaskVisualizer
 
-- When using Flash Attention 2 via `attn_implementation="flash_attention_2"`, don't pass `torch_dtype` to the `from_pretrained` class method and use Automatic Mixed-Precision training. When using `Trainer`, it is simply specifying either `fp16` or `bf16` to `True`. Otherwise, make sure you are using `torch.autocast`. This is required because the Flash Attention only support `fp16` and `bf16` data type.
+visualizer = AttentionMaskVisualizer("meta-llama/Llama-2-7b-hf")
+visualizer("Plants create energy through a process known as")
+```
 
+<div class="flex justify-center">
+    <img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/llama-2-attn-mask.png"/>
+</div>
 
-## Resources
+## Notes
 
-A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with LLaMA2. If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.
+- Setting `config.pretraining_tp` to a value besides `1` activates a more accurate but slower computation of the linear layers. This matches the original logits better.
+- The original model uses `pad_id = -1` to indicate a padding token. The Transformers implementation requires adding a padding token and resizing the token embedding accordingly.
 
-- [Llama 2 is here - get it on Hugging Face](https://huggingface.co/blog/llama2), a blog post about Llama 2 and how to use it with 🤗 Transformers and 🤗 PEFT.
-- [LLaMA 2 - Every Resource you need](https://www.philschmid.de/llama-2), a compilation of relevant resources to learn about LLaMA 2 and how to get started quickly.
-
-<PipelineTag pipeline="text-generation"/>
-
-- A [notebook](https://colab.research.google.com/drive/1PEQyJO1-f6j0S_XJ8DV50NkpzasXkrzd?usp=sharing) on how to fine-tune Llama 2 in Google Colab using QLoRA and 4-bit precision. 🌎
-- A [notebook](https://colab.research.google.com/drive/134o_cXcMe_lsvl15ZE_4Y75Kstepsntu?usp=sharing) on how to fine-tune the "Llama-v2-7b-guanaco" model with 4-bit QLoRA and generate Q&A datasets from PDFs. 🌎
-
-<PipelineTag pipeline="text-classification"/>
-
-- A [notebook](https://colab.research.google.com/drive/1ggaa2oRFphdBmqIjSEbnb_HGkcIRC2ZB?usp=sharing) on how to fine-tune the Llama 2 model with QLoRa, TRL, and Korean text classification dataset. 🌎🇰🇷
-
-⚗️ Optimization
-- [Fine-tune Llama 2 with DPO](https://huggingface.co/blog/dpo-trl), a guide to using the TRL library's DPO method to fine tune Llama 2 on a specific dataset.
-- [Extended Guide: Instruction-tune Llama 2](https://www.philschmid.de/instruction-tune-llama-2), a guide to training Llama 2 to generate instructions from inputs, transforming the model from instruction-following to instruction-giving.
-- A [notebook](https://colab.research.google.com/drive/1SYpgFpcmtIUzdE7pxqknrM4ArCASfkFQ?usp=sharing) on how to fine-tune the Llama 2 model on a personal computer using QLoRa and TRL. 🌎
-
-⚡️ Inference
-- A [notebook](https://colab.research.google.com/drive/1TC56ArKerXUpbgRy5vM3woRsbTEVNq7h?usp=sharing) on how to quantize the Llama 2 model using GPTQ from the AutoGPTQ library. 🌎
-- A [notebook](https://colab.research.google.com/drive/1X1z9Q6domMKl2CnEM0QGHNwidLfR4dW2?usp=sharing) on how to run the Llama 2 Chat Model with 4-bit quantization on a local computer or Google Colab. 🌎
-
-🚀 Deploy
-- [Fine-tune LLaMA 2 (7-70B) on Amazon SageMaker](https://www.philschmid.de/sagemaker-llama2-qlora), a complete guide from setup to QLoRA fine-tuning and deployment on Amazon SageMaker.
-- [Deploy Llama 2 7B/13B/70B on Amazon SageMaker](https://www.philschmid.de/sagemaker-llama-llm), a guide on using Hugging Face's LLM DLC container for secure and scalable deployment.
+    ```py
+    tokenizer.add_special_tokens({"pad_token":"<pad>"})
+    # update model config with padding token
+    model.config.pad_token_id
+    ```
+- It is recommended to initialize the `embed_tokens` layer with the following code to ensure encoding the padding token outputs zeros.
 
+    ```py
+    self.embed_tokens = nn.Embedding(config.vocab_size, config.hidden_size, self.config.padding_idx)
+    ```
+- The tokenizer is a byte-pair encoding model based on [SentencePiece](https://github.com/google/sentencepiece). During decoding, if the first token is the start of the word (for example, "Banana"), the tokenizer doesn't prepend the prefix space to the string.
+- Don't use the `torch_dtype` parameter in [`~AutoModel.from_pretrained`] if you're using FlashAttention-2 because it only supports fp16 or bf16. You should use [Automatic Mixed Precision](https://pytorch.org/tutorials/recipes/recipes/amp_recipe.html), set fp16 or bf16 to `True` if using [`Trainer`], or use [torch.autocast](https://pytorch.org/docs/stable/amp.html#torch.autocast).
 
 ## LlamaConfig
 

docs/source/en/model_doc/modernbert.md

@@ -60,6 +60,9 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
 
 - [Masked language modeling task guide](../tasks/masked_language_modeling)
 
+<PipelineTag pipeline="question-answering"/>
+
+- [`ModernBertForQuestionAnswering`] is supported by this [example script](https://github.com/huggingface/transformers/tree/main/examples/pytorch/question-answering) and [colab notebook](https://colab.research.google.com/github/huggingface/notebooks/blob/main/examples/question_answering.ipynb).
 
 ## ModernBertConfig
 
@@ -88,5 +91,15 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
 [[autodoc]] ModernBertForTokenClassification
     - forward
 
+## ModernBertForQuestionAnswering
+
+[[autodoc]] ModernBertForQuestionAnswering
+    - forward
+
+### Usage tips
+
+The ModernBert model can be fine-tuned using the HuggingFace Transformers library with its [official script](https://github.com/huggingface/transformers/blob/main/examples/pytorch/question-answering/run_qa.py) for question-answering tasks.
+
+
 </pt>
 </frameworkcontent>

docs/source/en/model_doc/paligemma.md

@@ -14,89 +14,157 @@ rendered properly in your Markdown viewer.
 
 -->
 
-# PaliGemma
-
-<div class="flex flex-wrap space-x-1">
-<img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
-<img alt="FlashAttention" src="https://img.shields.io/badge/%E2%9A%A1%EF%B8%8E%20FlashAttention-eae0c8?style=flat">
-<img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+<div style="float: right;">
+    <div class="flex flex-wrap space-x-1">
+        <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
+        <img alt="FlashAttention" src="https://img.shields.io/badge/%E2%9A%A1%EF%B8%8E%20FlashAttention-eae0c8?style=flat">
+        <img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+    </div>
 </div>
 
-## Overview
+# PaliGemma
 
-The PaliGemma model was proposed in [PaliGemma – Google's Cutting-Edge Open Vision Language Model](https://huggingface.co/blog/paligemma) by Google. It is a 3B vision-language model composed by a [SigLIP](siglip) vision encoder and a [Gemma](gemma) language decoder linked by a multimodal linear projection. It cuts an image into a fixed number of VIT tokens and prepends it to an optional prompt. One particularity is that the model uses full block attention on all the image tokens plus the input text tokens. It comes in 3 resolutions, 224x224, 448x448 and 896x896 with 3 base models, with 55 fine-tuned versions for different tasks, and 2 mix models.
+[PaliGemma](https://huggingface.co/papers/2407.07726) is a family of vision-language models (VLMs), combining [SigLIP](./siglip) with the [Gemma](./gemma) 2B model. PaliGemma is available in 3B, 10B, and 28B parameters. The main purpose of PaliGemma is to provide an adaptable base VLM that is easy to transfer to other tasks. The SigLIP vision encoder is a "shape optimized" contrastively pretrained [ViT](./vit) that converts an image into a sequence of tokens and prepended to an optional prompt. The Gemma 2B model is used as the decoder. PaliGemma uses full attention on all image and text tokens to maximize its capacity.
 
-<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/blog/paligemma/paligemma_arch.png"
-alt="drawing" width="600"/>
+[PaliGemma 2](https://huggingface.co/papers/2412.03555) improves on the first model by using Gemma 2 (2B, 9B, and 27B parameter variants) as the decoder. These are available as **pt** or **mix** variants. The **pt** checkpoints are intended for further fine-tuning and the **mix** checkpoints are ready for use out of the box.
 
-<small> PaliGemma architecture. Taken from the <a href="https://huggingface.co/blog/paligemma">blog post.</a> </small>
+You can find all the original PaliGemma checkpoints under the [PaliGemma](https://huggingface.co/collections/google/paligemma-release-6643a9ffbf57de2ae0448dda), [PaliGemma 2](https://huggingface.co/collections/google/paligemma-2-release-67500e1e1dbfdd4dee27ba48), and [PaliGemma 2 Mix](https://huggingface.co/collections/google/paligemma-2-mix-67ac6a251aaf3ee73679dcc4) collections.
 
-This model was contributed by [Molbap](https://huggingface.co/Molbap).
+> [!TIP]
+> Click on the PaliGemma models in the right sidebar for more examples of how to apply PaliGemma to different vision and language tasks.
 
-## Usage tips
+The example below demonstrates how to generate text based on an image with [`Pipeline`] or the [`AutoModel`] class.
 
-- PaliGemma is not meant for conversational use, and it works best when fine-tuning to a specific use case. Some downstream tasks on which PaliGemma can be fine-tuned include image captioning, visual question answering (VQA), object detection, referring expression segmentation and document understanding.
-- One can use `PaliGemmaProcessor` to prepare images, text and optional labels for the model. When fine-tuning a PaliGemma model, the `suffix` argument can be passed to the processor which creates the `labels` for the model:
+<hfoptions id="usage">
+<hfoption id="Pipeline">
 
-```python
-prompt = "What is on the flower?"
-answer = "a bee"
-inputs = processor(images=raw_image, text=prompt, suffix=answer, return_tensors="pt")
+```py
+import torch
+from transformers import pipeline
+
+pipeline = pipeline(
+    task="image-text-to-text",
+    model="google/paligemma2-3b-mix-224",
+    device=0,
+    torch_dtype=torch.bfloat16
+)
+pipeline(
+    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg",
+    text="What is in this image?"
+)
 ```
 
-## Usage Example
+</hfoption>
+<hfoption id="AutoModel">
 
-The model can accept a single or multiple images. According to the [paper](https://arxiv.org/abs/2407.07726v1), the checkpoint PaliGemma can transfer to tasks which take multiple images as input. NLVR2 is one such task, which asks one question about two images, and requires looking at both to give the correct answer. Here's an example code for single and multi image inference.
-
-### Single-image Inference
-
-```python
+```py
+import torch
+import requests
+from PIL import Image
 from transformers import AutoProcessor, PaliGemmaForConditionalGeneration
 
-model_id = "google/paligemma-3b-mix-224"
-model = PaliGemmaForConditionalGeneration.from_pretrained(model_id)
-processor = AutoProcessor.from_pretrained(model_id)
-
-prompt = "What is on the flower?"
-image_file = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg?download=true"
-raw_image = Image.open(requests.get(image_file, stream=True).raw)
-inputs = processor(raw_image, prompt, return_tensors="pt")
-output = model.generate(**inputs, max_new_tokens=20)
-
-print(processor.decode(output[0], skip_special_tokens=True)[inputs.input_ids.shape[1]: ])
-```
-
-### Multi-image Inference
-
-```python
-model_id = "google/paligemma-3b-ft-nlvr2-448"  # checkpoint tuned for multiple images
-model = PaliGemmaForConditionalGeneration.from_pretrained(model_id)
-processor = PaliGemmaProcessor.from_pretrained(model_id)
-
-prompt = "answer en Which of the two pictures shows a snowman, first or second?"
-stop_sign_image = Image.open(
-    requests.get("https://www.ilankelman.org/stopsigns/australia.jpg", stream=True).raw
+model = PaliGemmaForConditionalGeneration.from_pretrained(
+    "google/paligemma2-3b-mix-224",
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    attn_implementation="sdpa"
 )
-snow_image = Image.open(
-    requests.get(
-        "https://huggingface.co/microsoft/kosmos-2-patch14-224/resolve/main/snowman.jpg", stream=True
-    ).raw
+processor = AutoProcessor.from_pretrained(
+    "google/paligemma2-3b-mix-224",
 )
 
-inputs = processor(images=[[snow_image, stop_sign_image]], text=prompt, return_tensors="pt")
-
-output = model.generate(**inputs, max_new_tokens=20)
-print(processor.decode(output[0], skip_special_tokens=True)[inputs.input_ids.shape[1]: ])
+prompt = "What is in this image?"
+url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"
+image = Image.open(requests.get(url, stream=True).raw)
+inputs = processor(image, prompt, return_tensors="pt").to("cuda")
 
+output = model.generate(**inputs, max_new_tokens=50, cache_implementation="static")
+print(processor.decode(output[0], skip_special_tokens=True))
 ```
 
-## Resources
+</hfoption>
+</hfoptions>
 
-A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with PaliGemma. If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.
+Quantization reduces the memory burden of large models by representing the weights in a lower precision. Refer to the [Quantization](../quantization/overview) overview for more available quantization backends.
 
-- A blog post introducing all the features of PaliGemma can be found [here](https://huggingface.co/blog/paligemma).
-- Demo notebooks on how to fine-tune PaliGemma for VQA with the Trainer API along with inference can be found [here](https://github.com/huggingface/notebooks/tree/main/examples/paligemma).
-- Demo notebooks on how to fine-tune PaliGemma on a custom dataset (receipt image -> JSON) along with inference can be found [here](https://github.com/NielsRogge/Transformers-Tutorials/tree/master/PaliGemma). 🌎
+The example below uses [torchao](../quantization/torchao) to only quantize the weights to int4.
+
+```py
+# pip install torchao
+import torch
+import requests
+from PIL import Image
+from transformers import TorchAoConfig, AutoProcessor, PaliGemmaForConditionalGeneration
+
+quantization_config = TorchAoConfig("int4_weight_only", group_size=128)
+model = PaliGemmaForConditionalGeneration.from_pretrained(
+    "google/paligemma2-28b-mix-224",
+    torch_dtype=torch.bfloat16,
+    device_map="auto",
+    quantization_config=quantization_config
+)
+processor = AutoProcessor.from_pretrained(
+    "google/paligemma2-28b-mix-224",
+)
+
+prompt = "What is in this image?"
+url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"
+image = Image.open(requests.get(url, stream=True).raw)
+inputs = processor(image, prompt, return_tensors="pt").to("cuda")
+
+output = model.generate(**inputs, max_new_tokens=50, cache_implementation="static")
+print(processor.decode(output[0], skip_special_tokens=True))
+```
+
+Use the [AttentionMaskVisualizer](https://github.com/huggingface/transformers/blob/beb9b5b02246b9b7ee81ddf938f93f44cfeaad19/src/transformers/utils/attention_visualizer.py#L139) to better understand what tokens the model can and cannot attend to.
+
+```py
+from transformers.utils.attention_visualizer import AttentionMaskVisualizer
+
+visualizer = AttentionMaskVisualizer("google/paligemma2-3b-mix-224")
+visualizer("<img> What is in this image?")
+```
+
+<div class="flex justify-center">
+    <img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/paligemma2-attn-mask.png"/>
+</div>
+
+## Notes
+
+- PaliGemma is not a conversational model and works best when fine-tuned for specific downstream tasks such as image captioning, visual question answering (VQA), object detection, and document understanding.
+- [`PaliGemmaProcessor`] can prepare images, text, and optional labels for the model. Pass the `suffix` parameter to the processor to create labels for the model during fine-tuning.
+
+    ```py
+    prompt = "What is in this image?"
+    answer = "a pallas cat"
+    inputs = processor(images=image, text=prompt, suffix=answer, return_tensors="pt")
+    ```
+- PaliGemma can support multiple input images if it is fine-tuned to accept multiple images. For example, the [NLVR2](https://huggingface.co/google/paligemma-3b-ft-nlvr2-448) checkpoint supports multiple images. Pass the images as a list to the processor.
+
+    ```py
+    import torch
+    import requests
+    from PIL import Image
+    from transformers import TorchAoConfig, AutoProcessor, PaliGemmaForConditionalGeneration
+
+    model = PaliGemmaForConditionalGeneration.from_pretrained("google/paligemma-3b-ft-nlvr2-448")
+    processor = AutoProcessor.from_pretrained("google/paligemma-3b-ft-nlvr2-448")
+
+    prompt = "Are these two images the same?"
+    cat_image = Image.open(
+        requests.get("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg", stream=True).raw
+    )
+    cow_image = Image.open(
+        requests.get(
+            "https://media.istockphoto.com/id/1192867753/photo/cow-in-berchida-beach-siniscola.jpg?s=612x612&w=0&k=20&c=v0hjjniwsMNfJSuKWZuIn8pssmD5h5bSN1peBd1CmH4=", stream=True
+        ).raw
+    )
+
+    inputs = processor(images=[[cat_image, cow_image]], text=prompt, return_tensors="pt")
+
+    output = model.generate(**inputs, max_new_tokens=20, cache_implementation="static")
+    print(processor.decode(output[0], skip_special_tokens=True))
+    ```
 
 ## PaliGemmaConfig
 

docs/source/en/model_doc/phi4_multimodal.md

@@ -0,0 +1,149 @@
+<!--Copyright 2025 The HuggingFace Team. All rights reserved.
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+http://www.apache.org/licenses/LICENSE-2.0
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+rendered properly in your Markdown viewer.
+-->
+
+# Phi4 Multimodal
+
+## Overview
+
+Phi4 Multimodal is a lightweight open multimodal foundation model that leverages the language, vision, and speech research and datasets used for Phi-3.5 and 4.0 models. The model processes text, image, and audio inputs, generating text outputs, and comes with 128K token context length. The model underwent an enhancement process, incorporating both supervised fine-tuning, direct preference optimization and RLHF (Reinforcement Learning from Human Feedback) to support precise instruction adherence and safety measures. The languages that each modal supports are the following:
+
+- Text: Arabic, Chinese, Czech, Danish, Dutch, English, Finnish, French, German, Hebrew, Hungarian, Italian, Japanese, Korean, Norwegian, Polish, Portuguese, Russian, Spanish, Swedish, Thai, Turkish, Ukrainian
+- Vision: English
+- Audio: English, Chinese, German, French, Italian, Japanese, Spanish, Portuguese
+
+This model was contributed by [Cyril Vallez](https://huggingface.co/cyrilvallez). The most recent code can be
+found [here](https://github.com/huggingface/transformers/blob/main/src/transformers/models/phi4_multimodal/modeling_phi4_multimodal.py).
+
+
+## Usage tips
+
+`Phi4-multimodal-instruct` can be found on the [Huggingface Hub](https://huggingface.co/microsoft/Phi-4-multimodal-instruct)
+
+In the following, we demonstrate how to use it for inference depending on the input modalities (text, image, audio).
+
+```python
+import requests
+import torch
+import os
+import io
+from PIL import Image
+import soundfile as sf
+from transformers import AutoModelForCausalLM, AutoProcessor, GenerationConfig
+from urllib.request import urlopen
+
+
+# Define model path
+model_path = "microsoft/Phi-4-multimodal-instruct"
+device = "cuda:0"
+
+# Load model and processor
+processor = AutoProcessor.from_pretrained(model_path)
+model = AutoModelForCausalLM.from_pretrained(model_path, device_map=device,  torch_dtype=torch.float16)
+
+# Optional: load the adapters (note that without them, the base model will very likely not work well)
+model.load_adapter(model_path, adapter_name="speech", device_map=device, adapter_kwargs={"subfolder": 'speech-lora'})
+model.load_adapter(model_path, adapter_name="vision", device_map=device, adapter_kwargs={"subfolder": 'vision-lora'})
+
+# Define prompt structure
+user_prompt = '<|user|>'
+assistant_prompt = '<|assistant|>'
+prompt_suffix = '<|end|>'
+
+# Part 1: Image Processing
+model.set_adapter("vision") # if loaded, activate the vision adapter
+print("\n--- IMAGE PROCESSING ---")
+image_url = 'https://www.ilankelman.org/stopsigns/australia.jpg'
+prompt = f'{user_prompt}<|image_1|>What is shown in this image?{prompt_suffix}{assistant_prompt}'
+print(f'>>> Prompt\n{prompt}')
+
+# Download and open image
+image = Image.open(requests.get(image_url, stream=True).raw)
+inputs = processor(text=prompt, images=image, return_tensors='pt').to(device)
+
+# Generate response
+generate_ids = model.generate(
+    **inputs,
+    max_new_tokens=1000,
+    do_sample=False,
+)
+generate_ids = generate_ids[:, inputs['input_ids'].shape[1]:]
+response = processor.batch_decode(
+    generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False
+)[0]
+print(f'>>> Response\n{response}')
+
+# Part 2: Audio Processing
+model.set_adapter("speech") # if loaded, activate the speech adapter
+print("\n--- AUDIO PROCESSING ---")
+audio_url = "https://upload.wikimedia.org/wikipedia/commons/b/b0/Barbara_Sahakian_BBC_Radio4_The_Life_Scientific_29_May_2012_b01j5j24.flac"
+speech_prompt = "Transcribe the audio to text, and then translate the audio to French. Use <sep> as a separator between the original transcript and the translation."
+prompt = f'{user_prompt}<|audio_1|>{speech_prompt}{prompt_suffix}{assistant_prompt}'
+print(f'>>> Prompt\n{prompt}')
+
+# Downlowd and open audio file
+audio, sample_rate = sf.read(io.BytesIO(urlopen(audio_url).read()))
+
+# Process with the model
+inputs = processor(text=prompt, audios=audio, sample_rate=sample_rate, return_tensors='pt').to(device)
+
+generate_ids = model.generate(
+    **inputs,
+    max_new_tokens=1000,
+    do_sample=False,
+)
+generate_ids = generate_ids[:, inputs['input_ids'].shape[1]:]
+response = processor.batch_decode(
+    generate_ids, skip_special_tokens=True, clean_up_tokenization_spaces=False
+)[0]
+print(f'>>> Response\n{response}')
+```
+
+## Phi4MultimodalFeatureExtractor
+
+[[autodoc]] Phi4MultimodalFeatureExtractor
+
+## Phi4MultimodalImageProcessorFast
+
+[[autodoc]] Phi4MultimodalImageProcessorFast
+
+## Phi4MultimodalProcessor
+
+[[autodoc]] Phi4MultimodalProcessor
+
+## Phi4MultimodalAudioConfig
+
+[[autodoc]] Phi4MultimodalAudioConfig
+
+## Phi4MultimodalVisionConfig
+
+[[autodoc]] Phi4MultimodalVisionConfig
+
+## Phi4MultimodalConfig
+
+[[autodoc]] Phi4MultimodalConfig
+
+## Phi4MultimodalAudioModel
+
+[[autodoc]] Phi4MultimodalAudioModel
+
+## Phi4MultimodalVisionModel
+
+[[autodoc]] Phi4MultimodalVisionModel
+
+## Phi4MultimodalModel
+
+[[autodoc]] Phi4MultimodalModel
+    - forward
+
+## Phi4MultimodalForCausalLM
+
+[[autodoc]] Phi4MultimodalForCausalLM
+    - forward

docs/source/en/model_doc/qwen3.md

@@ -0,0 +1,59 @@
+<!--Copyright 2024 The Qwen Team and The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+
+⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+rendered properly in your Markdown viewer.
+
+-->
+
+# Qwen3
+
+## Overview
+
+To be released with the official model launch.
+
+### Model Details
+
+To be released with the official model launch.
+
+
+## Usage tips
+
+To be released with the official model launch.
+
+## Qwen3Config
+
+[[autodoc]] Qwen3Config
+
+## Qwen3Model
+
+[[autodoc]] Qwen3Model
+    - forward
+
+## Qwen3ForCausalLM
+
+[[autodoc]] Qwen3ForCausalLM
+    - forward
+
+## Qwen3ForSequenceClassification
+
+[[autodoc]] Qwen3ForSequenceClassification
+    - forward
+
+## Qwen3ForTokenClassification
+
+[[autodoc]] Qwen3ForTokenClassification
+    - forward
+
+## Qwen3ForQuestionAnswering
+
+[[autodoc]] Qwen3ForQuestionAnswering
+    - forward

docs/source/en/model_doc/qwen3_moe.md

@@ -0,0 +1,58 @@
+<!--Copyright 2024 The Qwen Team and The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+
+⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+rendered properly in your Markdown viewer.
+
+-->
+
+# Qwen3MoE
+
+## Overview
+
+To be released with the official model launch.
+
+### Model Details
+
+To be released with the official model launch.
+
+## Usage tips
+
+To be released with the official model launch.
+
+## Qwen3MoeConfig
+
+[[autodoc]] Qwen3MoeConfig
+
+## Qwen3MoeModel
+
+[[autodoc]] Qwen3MoeModel
+    - forward
+
+## Qwen3MoeForCausalLM
+
+[[autodoc]] Qwen3MoeForCausalLM
+    - forward
+
+## Qwen3MoeForSequenceClassification
+
+[[autodoc]] Qwen3MoeForSequenceClassification
+    - forward
+
+## Qwen3MoeForTokenClassification
+
+[[autodoc]] Qwen3MoeForTokenClassification
+    - forward
+
+## Qwen3MoeForQuestionAnswering
+
+[[autodoc]] Qwen3MoeForQuestionAnswering
+    - forward

docs/source/en/model_doc/vit.md

@@ -14,144 +14,83 @@ rendered properly in your Markdown viewer.
 
 -->
 
-# Vision Transformer (ViT)
-
-<div class="flex flex-wrap space-x-1">
-<img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
-<img alt="TensorFlow" src="https://img.shields.io/badge/TensorFlow-FF6F00?style=flat&logo=tensorflow&logoColor=white">
-<img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAC0AAAAtCAMAAAANxBKoAAAC7lBMVEUAAADg5vYHPVgAoJH+/v76+v39/f9JbLP///9+AIgAnY3///+mcqzt8fXy9fgkXa3Ax9709fr+///9/f8qXq49qp5AaLGMwrv8/P0eW60VWawxYq8yqJzG2dytt9Wyu9elzci519Lf3O3S2efY3OrY0+Xp7PT///////+dqNCexMc6Z7AGpJeGvbenstPZ5ejQ1OfJzOLa7ejh4+/r8fT29vpccbklWK8PVa0AS6ghW63O498vYa+lsdKz1NDRt9Kw1c672tbD3tnAxt7R6OHp5vDe7OrDyuDn6vLl6/EAQKak0MgATakkppo3ZK/Bz9y8w9yzu9jey97axdvHzeG21NHH4trTwthKZrVGZLSUSpuPQJiGAI+GAI8SWKydycLL4d7f2OTi1+S9xNzL0ePT6OLGzeEAo5U0qJw/aLEAo5JFa7JBabEAp5Y4qZ2QxLyKmsm3kL2xoMOehrRNb7RIbbOZgrGre68AUqwAqZqNN5aKJ5N/lMq+qsd8kMa4pcWzh7muhLMEV69juq2kbKqgUaOTR5uMMZWLLZSGAI5VAIdEAH+ovNDHuNCnxcy3qcaYx8K8msGplrx+wLahjbYdXrV6vbMvYK9DrZ8QrZ8tqJuFms+Sos6sw8ecy8RffsNVeMCvmb43aLltv7Q4Y7EZWK4QWa1gt6meZKUdr6GOAZVeA4xPAISyveLUwtivxtKTpNJ2jcqfvcltiMiwwcfAoMVxhL+Kx7xjdrqTe60tsaNQs6KaRKACrJ6UTZwkqpqTL5pkHY4AloSgsd2ptNXPvNOOncuxxsqFl8lmg8apt8FJcr9EbryGxLqlkrkrY7dRa7ZGZLQ5t6iXUZ6PPpgVpZeJCJFKAIGareTa0+KJod3H0deY2M+esM25usmYu8d2zsJOdcBVvrCLbqcAOaaHaKQAMaScWqKBXqCXMJ2RHpiLF5NmJZAdAHN2kta11dKu1M+DkcZLdb+Mcql3TppyRJdzQ5ZtNZNlIY+DF4+voCOQAAAAZ3RSTlMABAT+MEEJ/RH+/TP+Zlv+pUo6Ifz8+fco/fz6+evr39S9nJmOilQaF/7+/f38+smmoYp6b1T+/v7++vj189zU0tDJxsGzsrKSfv34+Pf27dDOysG9t6+n/vv6+vr59uzr1tG+tZ6Qg9Ym3QAABR5JREFUSMeNlVVUG1EQhpcuxEspXqS0SKEtxQp1d3d332STTRpIQhIISQgJhODu7lAoDoUCpe7u7u7+1puGpqnCPOyZvffbOXPm/PsP9JfQgyCC+tmTABTOcbxDz/heENS7/1F+9nhvkHePG0wNDLbGWwdXL+rbLWvpmZHXD8+gMfBjTh+aSe6Gnn7lwQIOTR0c8wfX3PWgv7avbdKwf/ZoBp1Gp/PvuvXW3vw5ib7emnTW4OR+3D4jB9vjNJ/7gNvfWWeH/TO/JyYrsiKCRjVEZA3UB+96kON+DxOQ/NLE8PE5iUYgIXjFnCOlxEQMaSGVxjg4gxOnEycGz8bptuNjVx08LscIgrzH3umcn+KKtiBIyvzOO2O99aAdR8cF19oZalnCtvREUw79tCd5sow1g1UKM6kXqUx4T8wsi3sTjJ3yzDmmhenLXLpo8u45eG5y4Vvbk6kkC4LLtJMowkSQxmk4ggVJEG+7c6QpHT8vvW9X7/o7+3ELmiJi2mEzZJiz8cT6TBlanBk70cB5GGIGC1gRDdZ00yADLW1FL6gqhtvNXNG5S9gdSrk4M1qu7JAsmYshzDS4peoMrU/gT7qQdqYGZaYhxZmVbGJAm/CS/HloWyhRUlknQ9KYcExTwS80d3VNOxUZJpITYyspl0LbhArhpZCD9cRWEQuhYkNGMHToQ/2Cs6swJlb39CsllxdXX6IUKh/H5jbnSsPKjgmoaFQ1f8wRLR0UnGE/RcDEjj2jXG1WVTwUs8+zxfcrVO+vSsuOpVKxCfYZiQ0/aPKuxQbQ8lIz+DClxC8u+snlcJ7Yr1z1JPqUH0V+GDXbOwAib931Y4Imaq0NTIXPXY+N5L18GJ37SVWu+hwXff8l72Ds9XuwYIBaXPq6Shm4l+Vl/5QiOlV+uTk6YR9PxKsI9xNJny31ygK1e+nIRC1N97EGkFPI+jCpiHe5PCEy7oWqWSwRrpOvhFzcbTWMbm3ZJAOn1rUKpYIt/lDhW/5RHHteeWFN60qo98YJuoq1nK3uW5AabyspC1BcIEpOhft+SZAShYoLSvnmSfnYADUERP5jJn2h5XtsgCRuhYQqAvwTwn33+YWEKUI72HX5AtfSAZDe8F2DtPPm77afhl0EkthzuCQU0BWApgQIH9+KB0JhopMM7bJrdTRoleM2JAVNMyPF+wdoaz+XJpGoVAQ7WXUkcV7gT3oUZyi/ISIJAVKhgNp+4b4veCFhYVJw4locdSjZCp9cPUhLF9EZ3KKzURepMEtCDPP3VcWFx4UIiZIklIpFNfHpdEafIF2aRmOcrUmjohbT2WUllbmRvgfbythbQO3222fpDJoufaQPncYYuqoGtUEsCJZL6/3PR5b4syeSjZMQG/T2maGANlXT2v8S4AULWaUkCxfLyW8iW4kdka+nEMjxpL2NCwsYNBp+Q61PF43zyDg9Bm9+3NNySn78jMZUUkumqE4Gp7JmFOdP1vc8PpRrzj9+wPinCy8K1PiJ4aYbnTYpCCbDkBSbzhu2QJ1Gd82t8jI8TH51+OzvXoWbnXUOBkNW+0mWFwGcGOUVpU81/n3TOHb5oMt2FgYGjzau0Nif0Ss7Q3XB33hjjQHjHA5E5aOyIQc8CBrLdQSs3j92VG+3nNEjbkbdbBr9zm04ruvw37vh0QKOdeGIkckc80fX3KH/h7PT4BOjgCty8VZ5ux1MoO5Cf5naca2LAsEgehI+drX8o/0Nu+W0m6K/I9gGPd/dfx/EN/wN62AhsBWuAAAAAElFTkSuQmCC
-">
-<img alt="FlashAttention" src="https://img.shields.io/badge/%E2%9A%A1%EF%B8%8E%20FlashAttention-eae0c8?style=flat">
-<img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+<div style="float: right;">
+    <div class="flex flex-wrap space-x-1">
+        <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
+        <img alt="TensorFlow" src="https://img.shields.io/badge/TensorFlow-FF6F00?style=flat&logo=tensorflow&logoColor=white">
+        <img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,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
+        ">
+        <img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+    </div>
 </div>
 
-## Overview
+# Vision Transformer (ViT)
 
-The Vision Transformer (ViT) model was proposed in [An Image is Worth 16x16 Words: Transformers for Image Recognition
-at Scale](https://arxiv.org/abs/2010.11929) by Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk
-Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob
-Uszkoreit, Neil Houlsby. It's the first paper that successfully trains a Transformer encoder on ImageNet, attaining
-very good results compared to familiar convolutional architectures.
+[Vision Transformer (ViT)](https://huggingface.co/papers/2010.11929) is a transformer adapted for computer vision tasks. An image is split into smaller fixed-sized patches which are treated as a sequence of tokens, similar to words for NLP tasks. ViT requires less resources to pretrain compared to convolutional architectures and its performance on large datasets can be transferred to smaller downstream tasks.
 
-The abstract from the paper is the following:
+You can find all the original ViT checkpoints under the [Google](https://huggingface.co/google?search_models=vit) organization.
 
-*While the Transformer architecture has become the de-facto standard for natural language processing tasks, its
-applications to computer vision remain limited. In vision, attention is either applied in conjunction with
-convolutional networks, or used to replace certain components of convolutional networks while keeping their overall
-structure in place. We show that this reliance on CNNs is not necessary and a pure transformer applied directly to
-sequences of image patches can perform very well on image classification tasks. When pre-trained on large amounts of
-data and transferred to multiple mid-sized or small image recognition benchmarks (ImageNet, CIFAR-100, VTAB, etc.),
-Vision Transformer (ViT) attains excellent results compared to state-of-the-art convolutional networks while requiring
-substantially fewer computational resources to train.*
+> [!TIP]
+> Click on the ViT models in the right sidebar for more examples of how to apply ViT to different computer vision tasks.
 
-<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/vit_architecture.jpg"
-alt="drawing" width="600"/>
+The example below demonstrates how to classify an image with [`Pipeline`] or the [`AutoModel`] class.
 
-<small> ViT architecture. Taken from the <a href="https://arxiv.org/abs/2010.11929">original paper.</a> </small>
+<hfoptions id="usage">
+<hfoption id="Pipeline">
 
-Following the original Vision Transformer, some follow-up works have been made:
+```py
+import torch
+from transformers import pipeline
 
-- [DeiT](deit) (Data-efficient Image Transformers) by Facebook AI. DeiT models are distilled vision transformers.
-  The authors of DeiT also released more efficiently trained ViT models, which you can directly plug into [`ViTModel`] or
-  [`ViTForImageClassification`]. There are 4 variants available (in 3 different sizes): *facebook/deit-tiny-patch16-224*,
-  *facebook/deit-small-patch16-224*, *facebook/deit-base-patch16-224* and *facebook/deit-base-patch16-384*. Note that one should
-  use [`DeiTImageProcessor`] in order to prepare images for the model.
-
-- [BEiT](beit) (BERT pre-training of Image Transformers) by Microsoft Research. BEiT models outperform supervised pre-trained
-  vision transformers using a self-supervised method inspired by BERT (masked image modeling) and based on a VQ-VAE.
-
-- DINO (a method for self-supervised training of Vision Transformers) by Facebook AI. Vision Transformers trained using
-  the DINO method show very interesting properties not seen with convolutional models. They are capable of segmenting
-  objects, without having ever been trained to do so. DINO checkpoints can be found on the [hub](https://huggingface.co/models?other=dino).
-
-- [MAE](vit_mae) (Masked Autoencoders) by Facebook AI. By pre-training Vision Transformers to reconstruct pixel values for a high portion
-  (75%) of masked patches (using an asymmetric encoder-decoder architecture), the authors show that this simple method outperforms
-  supervised pre-training after fine-tuning.
-
-This model was contributed by [nielsr](https://huggingface.co/nielsr). The original code (written in JAX) can be
-found [here](https://github.com/google-research/vision_transformer).
-
-Note that we converted the weights from Ross Wightman's [timm library](https://github.com/rwightman/pytorch-image-models),
-who already converted the weights from JAX to PyTorch. Credits go to him!
-
-## Usage tips
-
-- To feed images to the Transformer encoder, each image is split into a sequence of fixed-size non-overlapping patches,
-  which are then linearly embedded. A [CLS] token is added to serve as representation of an entire image, which can be
-  used for classification. The authors also add absolute position embeddings, and feed the resulting sequence of
-  vectors to a standard Transformer encoder.
-- As the Vision Transformer expects each image to be of the same size (resolution), one can use
-  [`ViTImageProcessor`] to resize (or rescale) and normalize images for the model.
-- Both the patch resolution and image resolution used during pre-training or fine-tuning are reflected in the name of
-  each checkpoint. For example, `google/vit-base-patch16-224` refers to a base-sized architecture with patch
-  resolution of 16x16 and fine-tuning resolution of 224x224. All checkpoints can be found on the [hub](https://huggingface.co/models?search=vit).
-- The available checkpoints are either (1) pre-trained on [ImageNet-21k](http://www.image-net.org/) (a collection of
-  14 million images and 21k classes) only, or (2) also fine-tuned on [ImageNet](http://www.image-net.org/challenges/LSVRC/2012/) (also referred to as ILSVRC 2012, a collection of 1.3 million
-  images and 1,000 classes).
-- The Vision Transformer was pre-trained using a resolution of 224x224. During fine-tuning, it is often beneficial to
-  use a higher resolution than pre-training [(Touvron et al., 2019)](https://arxiv.org/abs/1906.06423), [(Kolesnikov
-  et al., 2020)](https://arxiv.org/abs/1912.11370). In order to fine-tune at higher resolution, the authors perform
-  2D interpolation of the pre-trained position embeddings, according to their location in the original image.
-- The best results are obtained with supervised pre-training, which is not the case in NLP. The authors also performed
-  an experiment with a self-supervised pre-training objective, namely masked patched prediction (inspired by masked
-  language modeling). With this approach, the smaller ViT-B/16 model achieves 79.9% accuracy on ImageNet, a significant
-  improvement of 2% to training from scratch, but still 4% behind supervised pre-training.
-
-### Using Scaled Dot Product Attention (SDPA)
-
-PyTorch includes a native scaled dot-product attention (SDPA) operator as part of `torch.nn.functional`. This function 
-encompasses several implementations that can be applied depending on the inputs and the hardware in use. See the 
-[official documentation](https://pytorch.org/docs/stable/generated/torch.nn.functional.scaled_dot_product_attention.html) 
-or the [GPU Inference](https://huggingface.co/docs/transformers/main/en/perf_infer_gpu_one#pytorch-scaled-dot-product-attention)
-page for more information.
-
-SDPA is used by default for `torch>=2.1.1` when an implementation is available, but you may also set 
-`attn_implementation="sdpa"` in `from_pretrained()` to explicitly request SDPA to be used.
-
-```
-from transformers import ViTForImageClassification
-model = ViTForImageClassification.from_pretrained("google/vit-base-patch16-224", attn_implementation="sdpa", torch_dtype=torch.float16)
-...
+pipeline = pipeline(
+    task="image-classification",
+    model="google/vit-base-patch16-224",
+    torch_dtype=torch.float16,
+    device=0
+)
+pipeline(images="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg")
 ```
 
-For the best speedups, we recommend loading the model in half-precision (e.g. `torch.float16` or `torch.bfloat16`).
+</hfoption>
+<hfoption id="AutoModel">
 
-On a local benchmark (A100-40GB, PyTorch 2.3.0, OS Ubuntu 22.04) with `float32` and `google/vit-base-patch16-224` model, we saw the following speedups during inference.
+```py
+import torch
+import requests
+from PIL import Image
+from transformers import AutoModelForImageClassification, AutoImageProcessor
 
-|   Batch size |   Average inference time (ms), eager mode |   Average inference time (ms), sdpa model |   Speed up, Sdpa / Eager (x) |
-|--------------|-------------------------------------------|-------------------------------------------|------------------------------|
-|            1 |                                         7 |                                         6 |                      1.17 |
-|            2 |                                         8 |                                         6 |                      1.33 |
-|            4 |                                         8 |                                         6 |                      1.33 |
-|            8 |                                         8 |                                         6 |                      1.33 |
+image_processor = AutoImageProcessor.from_pretrained(
+    "google/vit-base-patch16-224",
+    use_fast=True,
+)
+model = AutoModelForImageClassification.from_pretrained(
+    "google/vit-base-patch16-224",
+    torch_dtype=torch.float16,
+    device_map="auto",
+    attn_implementation="sdpa"
+)
+url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"
+image = Image.open(requests.get(url, stream=True).raw)
+inputs = image_processor(image, return_tensors="pt").to("cuda")
 
-## Resources
+with torch.no_grad():
+  logits = model(**inputs).logits
+predicted_class_id = logits.argmax(dim=-1).item()
 
-Demo notebooks regarding inference as well as fine-tuning ViT on custom data can be found [here](https://github.com/NielsRogge/Transformers-Tutorials/tree/master/VisionTransformer).
-A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with ViT. If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.
+class_labels = model.config.id2label
+predicted_class_label = class_labels[predicted_class_id]
+print(f"The predicted class label is: {predicted_class_label}")
+```
 
-`ViTForImageClassification` is supported by:
-<PipelineTag pipeline="image-classification"/>
+</hfoption>
+</hfoptions>
 
-- A blog post on how to [Fine-Tune ViT for Image Classification with Hugging Face Transformers](https://huggingface.co/blog/fine-tune-vit)
-- A blog post on [Image Classification with Hugging Face Transformers and `Keras`](https://www.philschmid.de/image-classification-huggingface-transformers-keras)
-- A notebook on [Fine-tuning for Image Classification with Hugging Face Transformers](https://github.com/huggingface/notebooks/blob/main/examples/image_classification.ipynb)
-- A notebook on how to [Fine-tune the Vision Transformer on CIFAR-10 with the Hugging Face Trainer](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/VisionTransformer/Fine_tuning_the_Vision_Transformer_on_CIFAR_10_with_the_%F0%9F%A4%97_Trainer.ipynb)
-- A notebook on how to [Fine-tune the Vision Transformer on CIFAR-10 with PyTorch Lightning](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/VisionTransformer/Fine_tuning_the_Vision_Transformer_on_CIFAR_10_with_PyTorch_Lightning.ipynb)
+## Notes
 
-⚗️ Optimization
-
-- A blog post on how to [Accelerate Vision Transformer (ViT) with Quantization using Optimum](https://www.philschmid.de/optimizing-vision-transformer)
-
-⚡️ Inference
-
-- A notebook on [Quick demo: Vision Transformer (ViT) by Google Brain](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/VisionTransformer/Quick_demo_of_HuggingFace_version_of_Vision_Transformer_inference.ipynb)
-
-🚀 Deploy
-
-- A blog post on [Deploying Tensorflow Vision Models in Hugging Face with TF Serving](https://huggingface.co/blog/tf-serving-vision)
-- A blog post on [Deploying Hugging Face ViT on Vertex AI](https://huggingface.co/blog/deploy-vertex-ai)
-- A blog post on [Deploying Hugging Face ViT on Kubernetes with TF Serving](https://huggingface.co/blog/deploy-tfserving-kubernetes)
+- The best results are obtained with supervised pretraining, and during fine-tuning, it may be better to use images with a resolution higher than 224x224.
+- Use [`ViTImageProcessorFast`] to resize (or rescale) and normalize images to the expected size.
+- The patch and image resolution are reflected in the checkpoint name. For example, google/vit-base-patch16-224, is the **base-sized** architecture with a patch resolution of 16x16 and fine-tuning resolution of 224x224.
 
 ## ViTConfig
 
@@ -172,9 +111,6 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
 [[autodoc]] ViTImageProcessorFast
     - preprocess
 
-<frameworkcontent>
-<pt>
-
 ## ViTModel
 
 [[autodoc]] ViTModel
@@ -190,9 +126,6 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
 [[autodoc]] ViTForImageClassification
     - forward
 
-</pt>
-<tf>
-
 ## TFViTModel
 
 [[autodoc]] TFViTModel
@@ -203,9 +136,6 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
 [[autodoc]] TFViTForImageClassification
     - call
 
-</tf>
-<jax>
-
 ## FlaxVitModel
 
 [[autodoc]] FlaxViTModel
@@ -215,6 +145,3 @@ A list of official Hugging Face and community (indicated by 🌎) resources to h
 
 [[autodoc]] FlaxViTForImageClassification
     - __call__
-
-</jax>
-</frameworkcontent>

docs/source/en/model_doc/whisper.md

@@ -14,152 +14,86 @@ rendered properly in your Markdown viewer.
 
 -->
 
-# Whisper
 
-<div class="flex flex-wrap space-x-1">
-<img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
-<img alt="TensorFlow" src="https://img.shields.io/badge/TensorFlow-FF6F00?style=flat&logo=tensorflow&logoColor=white">
-<img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,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
-">
-<img alt="FlashAttention" src="https://img.shields.io/badge/%E2%9A%A1%EF%B8%8E%20FlashAttention-eae0c8?style=flat">
-<img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+<div style="float: right;">
+    <div class="flex flex-wrap space-x-1">
+        <img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
+        <img alt="TensorFlow" src="https://img.shields.io/badge/TensorFlow-FF6F00?style=flat&logo=tensorflow&logoColor=white">
+        <img alt="Flax" src="https://img.shields.io/badge/Flax-29a79b.svg?style=flat&logo=data:image/png;base64,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
+        ">
+        <img alt="FlashAttention" src="https://img.shields.io/badge/%E2%9A%A1%EF%B8%8E%20FlashAttention-eae0c8?style=flat">
+        <img alt="SDPA" src="https://img.shields.io/badge/SDPA-DE3412?style=flat&logo=pytorch&logoColor=white">
+    </div>
 </div>
 
-## Overview
+# Whisper
 
-The Whisper model was proposed in [Robust Speech Recognition via Large-Scale Weak Supervision](https://cdn.openai.com/papers/whisper.pdf) by Alec Radford, Jong Wook Kim, Tao Xu, Greg Brockman, Christine McLeavey, Ilya Sutskever.
+[Whisper](https://hf.co/papers/2212.04356) is a encoder-decoder (sequence-to-sequence) transformer pretrained on 680,000 hours of labeled audio data. This amount of pretraining data enables zero-shot performance on audio tasks in English and many other languages. The decoder allows Whisper to map the encoders learned speech representations to useful outputs, such as text, without additional fine-tuning. Whisper just works out of the box.
 
-The abstract from the paper is the following:
+You can find all the original Whisper checkpoints under the [Whisper](https://huggingface.co/collections/openai/whisper-release-6501bba2cf999715fd953013) collection.
 
-*We study the capabilities of speech processing systems trained simply to predict large amounts of transcripts of audio on the internet. When scaled to 680,000 hours of multilingual and multitask supervision, the resulting models generalize well to standard benchmarks and are often competitive with prior fully supervised results but in a zeroshot transfer setting without the need for any finetuning. When compared to humans, the models approach their accuracy and robustness. We are releasing models and inference code to serve as a foundation for further work on robust speech processing.*
+> [!TIP]
+> Click on the Whisper models in the right sidebar for more examples of how to apply Whisper to different audio tasks.
 
-This model was contributed by [Arthur Zucker](https://huggingface.co/ArthurZ). The Tensorflow version of this model was contributed by [amyeroberts](https://huggingface.co/amyeroberts).
-The original code can be found [here](https://github.com/openai/whisper).
+The example below demonstrates how to automatically transcribe speech into text with [`Pipeline`] or the [`AutoModel`] class.
 
-## Quick usage
-
-You can run Whisper in less than 4 lines of code and transcribe in less than a minute!
-
-```python
-# pip install transformers torch
+<hfoptions id="usage">
+<hfoption id="Pipeline">
 
+```py
 import torch
 from transformers import pipeline
 
-whisper = pipeline("automatic-speech-recognition", "openai/whisper-large-v3", torch_dtype=torch.float16, device="cuda:0")
-
-transcription = whisper("<audio_file.mp3>")
-
-print(transcription["text"])
+pipeline = pipeline(
+    task="automatic-speech-recognition",
+    model="openai/whisper-large-v3-turbo",
+    torch_dtype=torch.float16,
+    device=0
+)
+pipeline("https://huggingface.co/datasets/Narsil/asr_dummy/resolve/main/mlk.flac")
 ```
 
-Voila! You can swap the model with any [Whisper checkpoints](https://huggingface.co/models?other=whisper&sort=downloads) on the Hugging Face Hub with the same pipeline based on your needs.
+</hfoption>
+<hfoption id="AutoModel">
 
-Bonus: You can replace `"cuda"` with `"mps"` to make it seamlessly work on Macs.
+```py
+# pip install datasets
+import torch
+from datasets import load_dataset
+from transformers import AutoProcessor, WhisperForConditionalGeneration
 
-## Usage tips
+processor = AutoProcessor.from_pretrained(
+    "openai/whisper-large-v3-turbo",
+)
+model = WhisperForConditionalGeneration.from_pretrained(
+    "openai/whisper-large-v3-turbo",
+    torch_dtype=torch.float16,
+    device_map="auto",
+    attn_implementation="sdpa"
+).to("cuda")
 
-- The model usually performs well without requiring any finetuning.
-- The architecture follows a classic encoder-decoder architecture, which means that it relies on the [`~generation.GenerationMixin.generate`] function for inference.
-- One can use [`WhisperProcessor`] to prepare audio for the model, and decode the predicted ID's back into text.
+ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
+audio_sample = ds[0]["audio"]
 
-- To convert the model and the processor, we recommend using the following:
+input_features = processor(
+    audio_sample["array"],
+    sampling_rate=audio_sample["sampling_rate"],
+    return_tensors="pt"
+).input_features
+input_features = input_features.to("cuda", dtype=torch.float16)
 
-```bash
-python src/transformers/models/whisper/convert_openai_to_hf.py --checkpoint_path "" --pytorch_dump_folder_path "Arthur/whisper-3" --convert_preprocessor True
-```
-The script will automatically determine all necessary parameters from the OpenAI checkpoint. A `tiktoken` library needs to be installed
-to perform the conversion of the OpenAI tokenizer to the `tokenizers` version.
-
-## Inference
-
-Here is a step-by-step guide to transcribing an audio sample using a pre-trained Whisper model:
-
-```python
->>> from datasets import load_dataset
->>> from transformers import WhisperProcessor, WhisperForConditionalGeneration
-
->>> # Select an audio file and read it:
->>> ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
->>> audio_sample = ds[0]["audio"]
-
->>> # Load the Whisper model in Hugging Face format:
->>> processor = WhisperProcessor.from_pretrained("openai/whisper-tiny.en")
->>> model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-tiny.en")
-
->>> # Use the model and processor to transcribe the audio:
->>> input_features = processor(
-...     audio_sample["array"], sampling_rate=audio_sample["sampling_rate"], return_tensors="pt"
-... ).input_features
-
->>> # Generate token ids
->>> predicted_ids = model.generate(input_features)
-
->>> # Decode token ids to text
->>> transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
-
->>> transcription[0]
-' Mr. Quilter is the apostle of the middle classes, and we are glad to welcome his gospel.'
+predicted_ids = model.generate(input_features, cache_implementation="static")
+transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
+transcription[0]
 ```
 
-Whisper is compatible with the following optimisations for both short and long-form generation:
-- [PyTorch Scaled Dot Product Attention (SDPA)](../perf_infer_gpu_one#pytorch-scaled-dot-product-attention): flash attention and memory-efficient attention kernels. Enabled by default for `torch>=2.1.1`.
-- [Flash Attention 2](../perf_infer_gpu_one#flashattention-2): improved implementation of flash attention through better parallelism and work partitioning. 
-- [torch.compile](../llm_optims#static-kv-cache-and-torchcompile): JIT-compile the forward pass to dispatch to efficient fused kernels.
+</hfoption>
+</hfoptions>
 
-As an example, the following codesnippet enables SDPA and `torch.compile` for up to 5x faster inference:
+## Notes
 
-```python
->>> from datasets import load_dataset
->>> from transformers import WhisperProcessor, WhisperForConditionalGeneration
-
->>> # Select an audio file and read it:
->>> ds = load_dataset("hf-internal-testing/librispeech_asr_dummy", "clean", split="validation")
->>> audio_sample = ds[0]["audio"]
-
->>> # Load the Whisper model with SDPA attention
->>> processor = WhisperProcessor.from_pretrained("openai/whisper-tiny.en")
->>> model = WhisperForConditionalGeneration.from_pretrained("openai/whisper-tiny.en", attn_implementation="sdpa")
-
->>> # Enable static cache and compile the forward pass
->>> model.generation_config.cache_implementation = "static"
->>> model.forward = torch.compile(model.forward, mode="reduce-overhead", fullgraph=True)
-
->>> # Use the model and processor to transcribe the audio:
->>> input_features = processor(
-...     audio_sample["array"], sampling_rate=audio_sample["sampling_rate"], return_tensors="pt"
-... ).input_features
-
->>> # Compile the forward pass
->>> for _ in range(2):
->>>     model.generate(input_features)
-
->>> # Generate token ids using compiled graph (fast!)
->>> predicted_ids = model.generate(input_features)
-
->>> # Decode token ids to text
->>> transcription = processor.batch_decode(predicted_ids, skip_special_tokens=True)
-
->>> transcription[0]
-' Mr. Quilter is the apostle of the middle classes, and we are glad to welcome his gospel.'
-```
-
-For more details on each optimisation, refer to the documentation linked above.
-
-## Resources
-
-A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with Whisper. If you're interested in submitting a resource to be included here, please feel free to open a Pull Request and we'll review it! The resource should ideally demonstrate something new instead of duplicating an existing resource.
-
-- [Fine-tune Whisper](https://huggingface.co/blog/fine-tune-whisper) on your own dataset for better downstream performance.
-- [Distil-Whisper](https://huggingface.co/distil-whisper): Upto 6x faster, 2x smaller distilled Whisper models for English. We release the [model checkpoints](https://huggingface.co/distil-whisper), and [distillation code](https://github.com/huggingface/distil-whisper).
-- A fork with a script to [convert a Whisper model in Hugging Face format to OpenAI format](https://github.com/zuazo-forks/transformers/blob/convert_hf_to_openai/src/transformers/models/whisper/convert_hf_to_openai.py). 🌎
-Usage example:
-```bash
-pip install -U openai-whisper
-python convert_hf_to_openai.py \
-    --checkpoint openai/whisper-tiny \
-    --whisper_dump_path whisper-tiny-openai.pt
-```
+- Whisper relies on [`~GenerationMixin.generate`] for inference.
+- The [`WhisperProcessor`] can be used for preparing audio and decoding predicted ids back into text.
 
 ## WhisperConfig
 
@@ -205,9 +139,6 @@ python convert_hf_to_openai.py \
     - batch_decode
     - decode
 
-<frameworkcontent>
-<pt>
-
 ## WhisperModel
 
 [[autodoc]] WhisperModel
@@ -230,9 +161,6 @@ python convert_hf_to_openai.py \
 [[autodoc]] WhisperForAudioClassification
     - forward
 
-</pt>
-<tf>
-
 ## TFWhisperModel
 
 [[autodoc]] TFWhisperModel
@@ -243,9 +171,6 @@ python convert_hf_to_openai.py \
 [[autodoc]] TFWhisperForConditionalGeneration
     - call
 
-</tf>
-<jax>
-
 ## FlaxWhisperModel
 
 [[autodoc]] FlaxWhisperModel
@@ -260,7 +185,3 @@ python convert_hf_to_openai.py \
 
 [[autodoc]] FlaxWhisperForAudioClassification
     - __call__
-
-</jax>
-</frameworkcontent>
-

docs/source/en/perf_infer_gpu_multi.md

@@ -44,11 +44,6 @@ import os
 import torch
 from transformers import AutoModelForCausalLM, AutoTokenizer
 
-# initialize distributed environment
-rank = int(os.environ["RANK"])
-device = torch.device(f"cuda:{rank}")
-torch.cuda.set_device(device)
-torch.distributed.init_process_group("nccl", device_id=device)
 
 # enable tensor parallelism
 model = AutoModelForCausalLM.from_pretrained(
@@ -59,7 +54,7 @@ model = AutoModelForCausalLM.from_pretrained(
 # prepare input tokens
 tokenizer = AutoTokenizer.from_pretrained("meta-llama/Meta-Llama-3-8B-Instruct")
 prompt = "Can I help"
-inputs = tokenizer(prompt, return_tensors="pt").input_ids.to(device)
+inputs = tokenizer(prompt, return_tensors="pt").input_ids.to(model.device)
 
 # distributed run
 outputs = model(inputs)
@@ -71,6 +66,13 @@ Launch the inference script above on [torchrun](https://pytorch.org/docs/stable/
 torchrun --nproc-per-node 4 demo.py
 ```
 
+For CPU, please binding different socket on each rank. For example, if you are using Intel 4th Gen Xeon:
+```bash
+export OMP_NUM_THREADS=56
+numactl -C 0-55 -m 0 torchrun --nnodes=2 --node_rank=0 --master_addr="127.0.0.1" --master_port=29500 --nproc-per-node 1 demo.py & numactl -C 56-111 -m 1 torchrun --nnodes=2 --node_rank=1 --master_addr="127.0.0.1" --master_port=29500 --nproc-per-node 1 demo.py & wait
+```
+The CPU benchmark data will be released soon.
+
 You can benefit from considerable speed ups for inference, especially for inputs with large batch size or long sequences.
 
 For a single forward pass on [Llama](./model_doc/llama) with a sequence length of 512 and various batch sizes, you can expect the following speed ups.

docs/source/en/processors.md

@@ -29,8 +29,8 @@ import requests
 
 processor = AutoProcessor.from_pretrained("google/paligemma-3b-pt-224")
 
-prompt = "answer en Where is the cow standing?"
-url = "https://huggingface.co/gv-hf/PaliGemma-test-224px-hf/resolve/main/cow_beach_1.png"
+prompt = "answer en Where is the cat standing?"
+url = "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/pipeline-cat-chonk.jpeg"
 image = Image.open(requests.get(url, stream=True).raw)
 
 inputs = processor(text=prompt, images=image, return_tensors="pt")

docs/source/es/quicktour.md

@@ -220,7 +220,7 @@ Pasa tu texto al tokenizador:
 El tokenizador devolverá un diccionario conteniendo:
 
 * [input_ids](./glossary#input-ids): representaciones numéricas de los tokens.
-* [atttention_mask](.glossary#attention-mask): indica cuáles tokens deben ser atendidos.
+* [attention_mask](.glossary#attention-mask): indica cuáles tokens deben ser atendidos.
 
 Como con el [`pipeline`], el tokenizador aceptará una lista de inputs. Además, el tokenizador también puede rellenar (pad, en inglés) y truncar el texto para devolver un lote (batch, en inglés) de longitud uniforme:
 

docs/source/it/perf_infer_cpu.md

@@ -23,7 +23,7 @@ Abbiamo integrato di recente `BetterTransformer` per fare inferenza più rapidam
 
 ## PyTorch JIT-mode (TorchScript)
 
-TorchScript è un modo di creare modelli serializzabili e ottimizzabili da codice PyTorch. Ogni programmma TorchScript può esere salvato da un processo Python  e caricato in un processo dove non ci sono dipendenze Python.
+TorchScript è un modo di creare modelli serializzabili e ottimizzabili da codice PyTorch. Ogni programma TorchScript può esere salvato da un processo Python  e caricato in un processo dove non ci sono dipendenze Python.
 Comparandolo con l'eager mode di default, jit mode in PyTorch normalmente fornisce prestazioni migliori per l'inferenza del modello da parte di metodologie di ottimizzazione come la operator fusion.
 
 Per una prima introduzione a TorchScript, vedi la Introduction to [PyTorch TorchScript tutorial](https://pytorch.org/tutorials/beginner/Intro_to_TorchScript_tutorial.html#tracing-modules).

docs/source/ko/_toctree.yml

@@ -720,6 +720,8 @@
         title: (번역중) Perceiver
       - local: in_translation
         title: (번역중) Pix2Struct
+      - local: model_doc/qwen2_vl
+        title: Qwen2VL
       - local: in_translation
         title: (번역중) Segment Anything
       - local: in_translation

docs/source/ko/model_doc/qwen2_vl.md

@@ -0,0 +1,303 @@
+<!--Copyright 2024 The HuggingFace Team. All rights reserved.
+
+Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
+the License. You may obtain a copy of the License at
+
+http://www.apache.org/licenses/LICENSE-2.0
+
+Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
+an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
+specific language governing permissions and limitations under the License.
+
+⚠️ Note that this file is in Markdown but contain specific syntax for our doc-builder (similar to MDX) that may not be
+rendered properly in your Markdown viewer.
+
+-->
+
+# Qwen2-VL[[Qwen2-VL]]
+
+<div class="flex flex-wrap space-x-1">
+<img alt="PyTorch" src="https://img.shields.io/badge/PyTorch-DE3412?style=flat&logo=pytorch&logoColor=white">
+<img alt="FlashAttention" src="https://img.shields.io/badge/%E2%9A%A1%EF%B8%8E%20FlashAttention-eae0c8?style=flat">
+</div>
+
+## Overview[[Overview]]
+
+[Qwen2-VL](https://qwenlm.github.io/blog/qwen2-vl/) 모델은 알리바바 리서치의 Qwen팀에서 개발한 [Qwen-VL](https://arxiv.org/pdf/2308.12966) 모델의 주요 업데이트 버전입니다.
+
+블로그의 요약은 다음과 같습니다:
+
+*이 블로그는 지난 몇 년간 Qwen-VL에서 중대한 개선을 거쳐 발전된 Qwen2-VL 모델을 소개합니다. 중요 개선 사항은 향상된 이미지 이해, 고급 비디오 이해, 통합 시각 에이전트 기능, 확장된 다언어 지원을 포함하고 있습니다.모델 아키텍처는 Naive Dynamic Resolution 지원을 통해 임의의 이미지 해상도를 처리할 수 있도록 최적화되었으며, 멀티모달 회전 위치 임베딩(M-ROPE)을 활용하여 1D 텍스트와 다차원 시각 데이터를 효과적으로 처리합니다. 이 업데이트된 모델은 시각 관련 작업에서 GPT-4o와 Claude 3.5 Sonnet 같은 선도적인 AI 시스템과 경쟁력 있는 성능을 보여주며, 텍스트 능력에서는 오픈소스 모델 중 상위권에 랭크되어 있습니다. 이러한 발전은 Qwen2-VL을 강력한 멀티모달 처리 및 추론 능력이 필요한 다양한 응용 분야에서 활용할 수 있는 다재다능한 도구로 만들어줍니다.*
+
+<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/qwen2_vl_architecture.jpeg"
+alt="drawing" width="600"/>
+
+<small> Qwen2-VL 구조. 출처: <a href="https://qwenlm.github.io/blog/qwen2-vl/">블로그 게시글</a> </small>
+
+이 모델은 [simonJJJ](https://huggingface.co/simonJJJ)에 의해 기여되었습니다.
+
+## 사용 예시[[Usage example]]
+
+### 단일 미디어 추론[[Single Media inference]]
+
+이 모델은 이미지와 비디오를 모두 인풋으로 받을 수 있습니다. 다음은 추론을 위한 예제 코드입니다.
+
+```python
+
+import torch
+from transformers import Qwen2VLForConditionalGeneration, AutoTokenizer, AutoProcessor
+
+# 사용 가능한 장치에서 모델을 반 정밀도(half-precision)로 로드
+model = Qwen2VLForConditionalGeneration.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", device_map="auto")
+processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct")
+
+
+conversation = [
+    {
+        "role":"user",
+        "content":[
+            {
+                "type":"image",
+                "url": "https://qianwen-res.oss-cn-beijing.aliyuncs.com/Qwen-VL/assets/demo.jpeg"
+            },
+            {
+                "type":"text",
+                "text":"Describe this image."
+            }
+        ]
+    }
+]
+
+inputs = processor.apply_chat_template(
+    conversation,
+    add_generation_prompt=True,
+    tokenize=True,
+    return_dict=True,
+    return_tensors="pt"
+).to(model.device)
+
+# 추론: 아웃풋 생성
+output_ids = model.generate(**inputs, max_new_tokens=128)
+generated_ids = [output_ids[len(input_ids):] for input_ids, output_ids in zip(inputs.input_ids, output_ids)]
+output_text = processor.batch_decode(generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True)
+print(output_text)
+
+
+
+# 비디오
+conversation = [
+    {
+        "role": "user",
+        "content": [
+            {"type": "video", "path": "/path/to/video.mp4"},
+            {"type": "text", "text": "What happened in the video?"},
+        ],
+    }
+]
+
+inputs = processor.apply_chat_template(
+    conversation,
+    video_fps=1,
+    add_generation_prompt=True,
+    tokenize=True,
+    return_dict=True,
+    return_tensors="pt"
+).to(model.device)
+
+
+# 추론: 아웃풋 생성
+output_ids = model.generate(**inputs, max_new_tokens=128)
+generated_ids = [output_ids[len(input_ids):] for input_ids, output_ids in zip(inputs.input_ids, output_ids)]
+output_text = processor.batch_decode(generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True)
+print(output_text)
+```
+
+### 배치 혼합 미디어 추론[[Batch Mixed Media Inference]]
+
+이 모델은 이미지, 비디오, 텍스트 등 다양한 유형의 데이터를 혼합하여 배치 입력으로 처리할 수 있습니다. 다음은 예제입니다.
+
+```python
+
+# 첫번째 이미지에 대한 대화
+conversation1 = [
+    {
+        "role": "user",
+        "content": [
+            {"type": "image", "path": "/path/to/image1.jpg"},
+            {"type": "text", "text": "Describe this image."}
+        ]
+    }
+]
+
+# 두 개의 이미지에 대한 대화
+conversation2 = [
+    {
+        "role": "user",
+        "content": [
+            {"type": "image", "path": "/path/to/image2.jpg"},
+            {"type": "image", "path": "/path/to/image3.jpg"},
+            {"type": "text", "text": "What is written in the pictures?"}
+        ]
+    }
+]
+
+# 순수 텍스트로만 이루어진 대화
+conversation3 = [
+    {
+        "role": "user",
+        "content": "who are you?"
+    }
+]
+
+
+# 혼합된 미디어로 이루어진 대화
+conversation4 = [
+    {
+        "role": "user",
+        "content": [
+            {"type": "image", "path": "/path/to/image3.jpg"},
+            {"type": "image", "path": "/path/to/image4.jpg"},
+            {"type": "video", "path": "/path/to/video.jpg"},
+            {"type": "text", "text": "What are the common elements in these medias?"},
+        ],
+    }
+]
+
+conversations = [conversation1, conversation2, conversation3, conversation4]
+# 배치 추론을 위한 준비
+ipnuts = processor.apply_chat_template(
+    conversations,
+    video_fps=1,
+    add_generation_prompt=True,
+    tokenize=True,
+    return_dict=True,
+    return_tensors="pt"
+).to(model.device)
+
+
+# 배치 추론
+output_ids = model.generate(**inputs, max_new_tokens=128)
+generated_ids = [output_ids[len(input_ids):] for input_ids, output_ids in zip(inputs.input_ids, output_ids)]
+output_text = processor.batch_decode(generated_ids, skip_special_tokens=True, clean_up_tokenization_spaces=True)
+print(output_text)
+```
+
+### 사용 팁[[Usage Tips]]
+
+#### 이미지 해상도 트레이드오프[[Image Resolution trade-off]]
+
+이 모델은 다양한 해상도의 입력을 지원합니다. 디폴트로 입력에 대해 네이티브(native) 해상도를 사용하지만, 더 높은 해상도를 적용하면 성능이 향상될 수 있습니다. 다만, 이는 더 많은 연산 비용을 초래합니다. 사용자는 최적의 설정을 위해 최소 및 최대 픽셀 수를 조정할 수 있습니다.
+
+```python
+min_pixels = 224*224
+max_pixels = 2048*2048
+processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", min_pixels=min_pixels, max_pixels=max_pixels)
+```
+
+제한된 GPU RAM의 경우, 다음과 같이 해상도를 줄일 수 있습니다:
+
+```python
+min_pixels = 256*28*28
+max_pixels = 1024*28*28 
+processor = AutoProcessor.from_pretrained("Qwen/Qwen2-VL-7B-Instruct", min_pixels=min_pixels, max_pixels=max_pixels)
+```
+이렇게 하면 각 이미지가 256~1024개의 토큰으로 인코딩됩니다. 여기서 28은 모델이 14 크기의 패치(patch)와 2의 시간 패치(temporal patch size)를 사용하기 때문에 나온 값입니다 (14 × 2 = 28).
+
+
+#### 다중 이미지 인풋[[Multiple Image Inputs]]
+
+기본적으로 이미지와 비디오 콘텐츠는 대화에 직접 포함됩니다. 여러 개의 이미지를 처리할 때는 이미지 및 비디오에 라벨을 추가하면 참조하기가 더 쉬워집니다. 사용자는 다음 설정을 통해 이 동작을 제어할 수 있습니다:
+
+```python
+conversation = [
+    {
+        "role": "user",
+        "content": [
+            {"type": "image"}, 
+            {"type": "text", "text": "Hello, how are you?"}
+        ]
+    },
+    {
+        "role": "assistant",
+        "content": "I'm doing well, thank you for asking. How can I assist you today?"
+    },
+    {
+        "role": "user",
+        "content": [
+            {"type": "text", "text": "Can you describe these images and video?"}, 
+            {"type": "image"}, 
+            {"type": "image"}, 
+            {"type": "video"}, 
+            {"type": "text", "text": "These are from my vacation."}
+        ]
+    },
+    {
+        "role": "assistant",
+        "content": "I'd be happy to describe the images and video for you. Could you please provide more context about your vacation?"
+    },
+    {
+        "role": "user",
+        "content": "It was a trip to the mountains. Can you see the details in the images and video?"
+    }
+]
+
+# 디폴트:
+prompt_without_id = processor.apply_chat_template(conversation, add_generation_prompt=True)
+# 예상 아웃풋: '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\n<|vision_start|><|image_pad|><|vision_end|>Hello, how are you?<|im_end|>\n<|im_start|>assistant\nI'm doing well, thank you for asking. How can I assist you today?<|im_end|>\n<|im_start|>user\nCan you describe these images and video?<|vision_start|><|image_pad|><|vision_end|><|vision_start|><|image_pad|><|vision_end|><|vision_start|><|video_pad|><|vision_end|>These are from my vacation.<|im_end|>\n<|im_start|>assistant\nI'd be happy to describe the images and video for you. Could you please provide more context about your vacation?<|im_end|>\n<|im_start|>user\nIt was a trip to the mountains. Can you see the details in the images and video?<|im_end|>\n<|im_start|>assistant\n'
+
+
+# id 추가
+prompt_with_id = processor.apply_chat_template(conversation, add_generation_prompt=True, add_vision_id=True)
+# 예상 아웃풋: '<|im_start|>system\nYou are a helpful assistant.<|im_end|>\n<|im_start|>user\nPicture 1: <|vision_start|><|image_pad|><|vision_end|>Hello, how are you?<|im_end|>\n<|im_start|>assistant\nI'm doing well, thank you for asking. How can I assist you today?<|im_end|>\n<|im_start|>user\nCan you describe these images and video?Picture 2: <|vision_start|><|image_pad|><|vision_end|>Picture 3: <|vision_start|><|image_pad|><|vision_end|>Video 1: <|vision_start|><|video_pad|><|vision_end|>These are from my vacation.<|im_end|>\n<|im_start|>assistant\nI'd be happy to describe the images and video for you. Could you please provide more context about your vacation?<|im_end|>\n<|im_start|>user\nIt was a trip to the mountains. Can you see the details in the images and video?<|im_end|>\n<|im_start|>assistant\n'
+
+```
+
+#### 빠른 생성을 위한 Flash-Attention 2[[Flash-Attention 2 to speed up generation]]
+
+첫번째로, Flash Attention 2의 최신 버전을 설치합니다:
+
+```bash
+pip install -U flash-attn --no-build-isolation
+```
+
+또한, Flash-Attention 2를 지원하는 하드웨어가 필요합니다. 자세한 내용은 공식 문서인 [flash attention repository](https://github.com/Dao-AILab/flash-attention)에서 확인할 수 있습니다. FlashAttention-2는 모델이 `torch.float16` 또는 `torch.bfloat16` 형식으로 로드된 경우에만 사용할 수 있습니다.
+
+Flash Attention-2를 사용하여 모델을 로드하고 실행하려면, 다음과 같이 모델을 로드할 때 `attn_implementation="flash_attention_2"` 옵션을 추가하면 됩니다:
+
+```python
+from transformers import Qwen2VLForConditionalGeneration
+
+model = Qwen2VLForConditionalGeneration.from_pretrained(
+    "Qwen/Qwen2-VL-7B-Instruct", 
+    torch_dtype=torch.bfloat16, 
+    attn_implementation="flash_attention_2",
+)
+```
+
+## Qwen2VLConfig
+
+[[autodoc]] Qwen2VLConfig
+
+## Qwen2VLImageProcessor
+
+[[autodoc]] Qwen2VLImageProcessor
+    - preprocess
+
+## Qwen2VLImageProcessorFast
+
+[[autodoc]] Qwen2VLImageProcessorFast
+    - preprocess
+
+## Qwen2VLProcessor
+
+[[autodoc]] Qwen2VLProcessor
+
+## Qwen2VLModel
+
+[[autodoc]] Qwen2VLModel
+    - forward
+
+## Qwen2VLForConditionalGeneration
+
+[[autodoc]] Qwen2VLForConditionalGeneration
+    - forward

docs/source/pt/quicktour.md

@@ -222,7 +222,7 @@ Passe o texto para o tokenizer:
 O tokenizer retornará um dicionário contendo:
 
 * [input_ids](./glossary#input-ids): representações numéricas de seus tokens.
-* [atttention_mask](.glossary#attention-mask): indica quais tokens devem ser atendidos.
+* [attention_mask](.glossary#attention-mask): indica quais tokens devem ser atendidos.
 
 Assim como o [`pipeline`], o tokenizer aceitará uma lista de entradas. Além disso, o tokenizer também pode preencher e truncar o texto para retornar um lote com comprimento uniforme:
 

examples/flax/language-modeling/run_bert_flax.py

@@ -53,4 +53,4 @@ for _ in range(nbenchmark):
     func()
 end = time.time()
 print(end - start)
-print(f"Throughput: {((nbenchmark * BS)/(end-start)):.3f} examples/sec")
+print(f"Throughput: {((nbenchmark * BS) / (end - start)):.3f} examples/sec")

examples/flax/question-answering/run_qa.py

@@ -61,7 +61,7 @@ from transformers.utils import check_min_version, send_example_telemetry
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 Array = Any
 Dataset = datasets.arrow_dataset.Dataset

examples/flax/speech-recognition/run_flax_speech_recognition_seq2seq.py

@@ -60,7 +60,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risk.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/flax/speech-recognition/requirements.txt")
 

examples/flax/text-classification/run_flax_glue.py

@@ -56,7 +56,7 @@ from transformers.utils import check_min_version, send_example_telemetry
 
 logger = logging.getLogger(__name__)
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 Array = Any
 Dataset = datasets.arrow_dataset.Dataset

examples/flax/token-classification/run_flax_ner.py

@@ -57,7 +57,7 @@ from transformers.utils.versions import require_version
 
 logger = logging.getLogger(__name__)
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/token-classification/requirements.txt")
 

examples/flax/vision/run_image_classification.py

@@ -36,12 +36,12 @@ import optax
 # for dataset and preprocessing
 import torch
 import torchvision
-import torchvision.transforms as transforms
 from flax import jax_utils
 from flax.jax_utils import pad_shard_unpad, unreplicate
 from flax.training import train_state
 from flax.training.common_utils import get_metrics, onehot, shard, shard_prng_key
 from huggingface_hub import HfApi
+from torchvision import transforms
 from tqdm import tqdm
 
 import transformers

examples/legacy/multiple_choice/utils_multiple_choice.py

@@ -113,7 +113,7 @@ if is_torch_available():
             with FileLock(lock_path):
                 if os.path.exists(cached_features_file) and not overwrite_cache:
                     logger.info(f"Loading features from cached file {cached_features_file}")
-                    self.features = torch.load(cached_features_file)
+                    self.features = torch.load(cached_features_file, weights_only=True)
                 else:
                     logger.info(f"Creating features from dataset file at {data_dir}")
                     label_list = processor.get_labels()

examples/legacy/pytorch-lightning/run_glue.py

@@ -81,7 +81,7 @@ class GLUETransformer(BaseTransformer):
 
         cached_features_file = self._feature_file(mode)
         logger.info("Loading features from cached file %s", cached_features_file)
-        features = torch.load(cached_features_file)
+        features = torch.load(cached_features_file, weights_only=True)
         all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
         all_attention_mask = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
         all_token_type_ids = torch.tensor([f.token_type_ids for f in features], dtype=torch.long)

examples/legacy/pytorch-lightning/run_ner.py

@@ -63,7 +63,7 @@ class NERTransformer(BaseTransformer):
             cached_features_file = self._feature_file(mode)
             if os.path.exists(cached_features_file) and not args.overwrite_cache:
                 logger.info("Loading features from cached file %s", cached_features_file)
-                features = torch.load(cached_features_file)
+                features = torch.load(cached_features_file, weights_only=True)
             else:
                 logger.info("Creating features from dataset file at %s", args.data_dir)
                 examples = self.token_classification_task.read_examples_from_file(args.data_dir, mode)
@@ -89,7 +89,7 @@ class NERTransformer(BaseTransformer):
         "Load datasets. Called after prepare data."
         cached_features_file = self._feature_file(mode)
         logger.info("Loading features from cached file %s", cached_features_file)
-        features = torch.load(cached_features_file)
+        features = torch.load(cached_features_file, weights_only=True)
         all_input_ids = torch.tensor([f.input_ids for f in features], dtype=torch.long)
         all_attention_mask = torch.tensor([f.attention_mask for f in features], dtype=torch.long)
         if features[0].token_type_ids is not None:

examples/legacy/question-answering/run_squad.py

@@ -105,8 +105,8 @@ def train(args, train_dataset, model, tokenizer):
         os.path.join(args.model_name_or_path, "scheduler.pt")
     ):
         # Load in optimizer and scheduler states
-        optimizer.load_state_dict(torch.load(os.path.join(args.model_name_or_path, "optimizer.pt")))
-        scheduler.load_state_dict(torch.load(os.path.join(args.model_name_or_path, "scheduler.pt")))
+        optimizer.load_state_dict(torch.load(os.path.join(args.model_name_or_path, "optimizer.pt"), weights_only=True))
+        scheduler.load_state_dict(torch.load(os.path.join(args.model_name_or_path, "scheduler.pt"), weights_only=True))
 
     if args.fp16:
         try:
@@ -417,7 +417,7 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
     # Init features and dataset from cache if it exists
     if os.path.exists(cached_features_file) and not args.overwrite_cache:
         logger.info("Loading features from cached file %s", cached_features_file)
-        features_and_dataset = torch.load(cached_features_file)
+        features_and_dataset = torch.load(cached_features_file, weights_only=True)
         features, dataset, examples = (
             features_and_dataset["features"],
             features_and_dataset["dataset"],

examples/legacy/run_swag.py

@@ -244,7 +244,7 @@ def load_and_cache_examples(args, tokenizer, evaluate=False, output_examples=Fal
     )
     if os.path.exists(cached_features_file) and not args.overwrite_cache and not output_examples:
         logger.info("Loading features from cached file %s", cached_features_file)
-        features = torch.load(cached_features_file)
+        features = torch.load(cached_features_file, weights_only=True)
     else:
         logger.info("Creating features from dataset file at %s", input_file)
         examples = read_swag_examples(input_file)

examples/legacy/seq2seq/README.md

@@ -209,7 +209,7 @@ th 56 \
 ```
 
 ### Multi-GPU Evaluation
-here is a command to run xsum evaluation on 8 GPUS. It is more than linearly faster than run_eval.py in some cases
+here is a command to run xsum evaluation on 8 GPUs. It is more than linearly faster than run_eval.py in some cases
 because it uses SortishSampler to minimize padding. You can also use it on 1 GPU. `data_dir` must have
 `{type_path}.source` and `{type_path}.target`. Run `./run_distributed_eval.py --help` for all clargs.
 

examples/legacy/seq2seq/convert_model_to_fp16.py

@@ -22,7 +22,7 @@ from tqdm import tqdm
 
 def convert(src_path: str, map_location: str = "cpu", save_path: Union[str, None] = None) -> None:
     """Convert a pytorch_model.bin or model.pt file to torch.float16 for faster downloads, less disk space."""
-    state_dict = torch.load(src_path, map_location=map_location)
+    state_dict = torch.load(src_path, map_location=map_location, weights_only=True)
     for k, v in tqdm(state_dict.items()):
         if not isinstance(v, torch.Tensor):
             raise TypeError("FP16 conversion only works on paths that are saved state dicts, like pytorch_model.bin")

examples/legacy/seq2seq/finetune_trainer.py

@@ -231,9 +231,9 @@ def main():
 
     # set decoder_start_token_id for MBart
     if model.config.decoder_start_token_id is None and isinstance(tokenizer, (MBartTokenizer, MBartTokenizerFast)):
-        assert (
-            data_args.tgt_lang is not None and data_args.src_lang is not None
-        ), "mBart requires --tgt_lang and --src_lang"
+        assert data_args.tgt_lang is not None and data_args.src_lang is not None, (
+            "mBart requires --tgt_lang and --src_lang"
+        )
         if isinstance(tokenizer, MBartTokenizer):
             model.config.decoder_start_token_id = tokenizer.lang_code_to_id[data_args.tgt_lang]
         else:

examples/legacy/seq2seq/run_eval_search.py

@@ -128,7 +128,7 @@ def run_search():
 
     results_sorted = sorted(results, key=operator.itemgetter(*task_score_names[task]), reverse=True)
     print(" | ".join([f"{col:{col_widths[col]}}" for col in col_names]))
-    print(" | ".join([f"{'-'*col_widths[col]}" for col in col_names]))
+    print(" | ".join([f"{'-' * col_widths[col]}" for col in col_names]))
     for row in results_sorted:
         print(" | ".join([f"{row[col]:{col_widths[col]}}" for col in col_names]))
 

examples/legacy/seq2seq/utils.py

@@ -282,9 +282,9 @@ class Seq2SeqDataCollator:
         self.tokenizer = tokenizer
         self.pad_token_id = tokenizer.pad_token_id
         self.decoder_start_token_id = decoder_start_token_id
-        assert (
-            self.pad_token_id is not None
-        ), f"pad_token_id is not defined for ({self.tokenizer.__class__.__name__}), it must be defined."
+        assert self.pad_token_id is not None, (
+            f"pad_token_id is not defined for ({self.tokenizer.__class__.__name__}), it must be defined."
+        )
         self.data_args = data_args
         self.tpu_num_cores = tpu_num_cores
         self.dataset_kwargs = {"add_prefix_space": True} if isinstance(tokenizer, BartTokenizer) else {}
@@ -593,7 +593,7 @@ def assert_all_frozen(model):
     model_grads: List[bool] = list(grad_status(model))
     n_require_grad = sum(lmap(int, model_grads))
     npars = len(model_grads)
-    assert not any(model_grads), f"{n_require_grad/npars:.1%} of {npars} weights require grad"
+    assert not any(model_grads), f"{n_require_grad / npars:.1%} of {npars} weights require grad"
 
 
 def assert_not_all_frozen(model):

examples/legacy/token-classification/tasks.py

@@ -131,7 +131,7 @@ class POS(TokenClassificationTask):
             s_p = preds_list[example_id]
             out = ""
             for token in sentence:
-                out += f'{token["form"]} ({token["upos"]}|{s_p.pop(0)}) '
+                out += f"{token['form']} ({token['upos']}|{s_p.pop(0)}) "
             out += "\n"
             writer.write(out)
             example_id += 1

examples/legacy/token-classification/utils_ner.py

@@ -242,7 +242,7 @@ if is_torch_available():
             with FileLock(lock_path):
                 if os.path.exists(cached_features_file) and not overwrite_cache:
                     logger.info(f"Loading features from cached file {cached_features_file}")
-                    self.features = torch.load(cached_features_file)
+                    self.features = torch.load(cached_features_file, weights_only=True)
                 else:
                     logger.info(f"Creating features from dataset file at {data_dir}")
                     examples = token_classification_task.read_examples_from_file(data_dir, mode)

examples/modular-transformers/modeling_dummy.py

@@ -4,6 +4,7 @@
 #             the file from the modular. If any change should be done, please apply the change to the
 #                          modular_dummy.py file directly. One of our CI enforces this.
 #                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+from functools import partial
 from typing import Callable, Optional, Tuple, Union
 
 import torch
@@ -544,7 +545,7 @@ class DummyModel(DummyPreTrainedModel):
 
             if self.gradient_checkpointing and self.training:
                 layer_outputs = self._gradient_checkpointing_func(
-                    decoder_layer.__call__,
+                    partial(decoder_layer.__call__, **flash_attn_kwargs),
                     hidden_states,
                     causal_mask,
                     position_ids,

examples/modular-transformers/modeling_multimodal1.py

@@ -4,6 +4,7 @@
 #             the file from the modular. If any change should be done, please apply the change to the
 #                          modular_multimodal1.py file directly. One of our CI enforces this.
 #                🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨🚨
+from functools import partial
 from typing import Callable, Optional, Tuple, Union
 
 import torch
@@ -544,7 +545,7 @@ class Multimodal1TextModel(Multimodal1TextPreTrainedModel):
 
             if self.gradient_checkpointing and self.training:
                 layer_outputs = self._gradient_checkpointing_func(
-                    decoder_layer.__call__,
+                    partial(decoder_layer.__call__, **flash_attn_kwargs),
                     hidden_states,
                     causal_mask,
                     position_ids,

examples/modular-transformers/modeling_multimodal2.py

@@ -534,7 +534,7 @@ class Multimodal2VisionEmbeddings(nn.Module):
         batch_size, _, height, width = pixel_values.shape
         if not interpolate_pos_encoding and (height != self.image_size or width != self.image_size):
             raise ValueError(
-                f"Input image size ({height}*{width}) doesn't match model" f" ({self.image_size}*{self.image_size})."
+                f"Input image size ({height}*{width}) doesn't match model ({self.image_size}*{self.image_size})."
             )
         target_dtype = self.patch_embedding.weight.dtype
         patch_embeds = self.patch_embedding(pixel_values.to(dtype=target_dtype))  # shape = [*, width, grid, grid]

examples/pytorch/audio-classification/run_audio_classification.py

@@ -45,7 +45,7 @@ from transformers.utils.versions import require_version
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=1.14.0", "To fix: pip install -r examples/pytorch/audio-classification/requirements.txt")
 

examples/pytorch/contrastive-image-text/run_clip.py

@@ -54,7 +54,7 @@ from transformers.utils.versions import require_version
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/contrastive-image-text/requirements.txt")
 

examples/pytorch/image-classification/run_image_classification.py

@@ -57,7 +57,7 @@ from transformers.utils.versions import require_version
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/pytorch/image-classification/requirements.txt")
 

examples/pytorch/image-classification/run_image_classification_no_trainer.py

@@ -49,7 +49,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 logger = get_logger(__name__)
 

examples/pytorch/image-pretraining/run_mae.py

@@ -43,7 +43,7 @@ from transformers.utils.versions import require_version
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/image-pretraining/requirements.txt")
 

examples/pytorch/image-pretraining/run_mim.py

@@ -48,7 +48,7 @@ Any model supported by the AutoModelForMaskedImageModeling API can be used.
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/image-pretraining/requirements.txt")
 

examples/pytorch/image-pretraining/run_mim_no_trainer.py

@@ -53,7 +53,7 @@ Any model supported by the AutoModelForMaskedImageModeling API can be used.
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/image-pretraining/requirements.txt")
 

examples/pytorch/instance-segmentation/README.md

@@ -66,7 +66,7 @@ python run_instance_segmentation.py \
     --dataloader_persistent_workers \
     --dataloader_prefetch_factor 4 \
     --do_eval \
-    --evaluation_strategy epoch \
+    --eval_strategy epoch \
     --logging_strategy epoch \
     --save_strategy epoch \
     --save_total_limit 2 \

examples/pytorch/instance-segmentation/run_instance_segmentation.py

@@ -46,7 +46,7 @@ from transformers.utils.versions import require_version
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=2.0.0", "To fix: pip install -r examples/pytorch/instance-segmentation/requirements.txt")
 

examples/pytorch/instance-segmentation/run_instance_segmentation_no_trainer.py

@@ -52,7 +52,7 @@ from transformers.utils.versions import require_version
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=2.0.0", "To fix: pip install -r examples/pytorch/instance-segmentation/requirements.txt")
 

examples/pytorch/language-modeling/run_clm.py

@@ -55,7 +55,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")
 
@@ -438,7 +438,7 @@ def main():
     else:
         model = AutoModelForCausalLM.from_config(config, trust_remote_code=model_args.trust_remote_code)
         n_params = sum({p.data_ptr(): p.numel() for p in model.parameters()}.values())
-        logger.info(f"Training new model from scratch - Total size={n_params/2**20:.2f}M params")
+        logger.info(f"Training new model from scratch - Total size={n_params / 2**20:.2f}M params")
 
     # We resize the embeddings only when necessary to avoid index errors. If you are creating a model from scratch
     # on a small vocab and want a smaller embedding size, remove this test.

examples/pytorch/language-modeling/run_clm_no_trainer.py

@@ -57,7 +57,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 logger = get_logger(__name__)
 

examples/pytorch/language-modeling/run_fim.py

@@ -47,7 +47,7 @@ from transformers import (
     Trainer,
     TrainingArguments,
     default_data_collator,
-    is_torch_tpu_available,
+    is_torch_xla_available,
     set_seed,
 )
 from transformers.integrations import is_deepspeed_zero3_enabled
@@ -58,7 +58,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")
 
@@ -265,8 +265,7 @@ class DataTrainingArguments:
         default="<fim_pad>",
         metadata={
             "help": (
-                "Fill-in-Middle Pad token. Used only when 'truncate_or_pad' is set to True. "
-                "Defaults to '<fim_pad>'."
+                "Fill-in-Middle Pad token. Used only when 'truncate_or_pad' is set to True. Defaults to '<fim_pad>'."
             )
         },
     )
@@ -514,7 +513,7 @@ def main():
             attn_implementation=model_args.attn_implementation,
         )
         n_params = sum({p.data_ptr(): p.numel() for p in model.parameters()}.values())
-        logger.info(f"Training new model from scratch - Total size={n_params/2**20:.2f}M params")
+        logger.info(f"Training new model from scratch - Total size={n_params / 2**20:.2f}M params")
 
     # Add the new FIM tokens to the tokenizer and resize model's vocab embeddings
     special_tokens = [data_args.fim_prefix_token, data_args.fim_middle_token, data_args.fim_suffix_token]
@@ -526,7 +525,7 @@ def main():
     if torch.cuda.is_availble():
         pad_factor = 8
 
-    elif is_torch_tpu_available():
+    elif is_torch_xla_available(check_is_tpu=True):
         pad_factor = 128
 
     # Add the new tokens to the tokenizer
@@ -796,9 +795,13 @@ def main():
         processing_class=tokenizer,
         # Data collator will default to DataCollatorWithPadding, so we change it.
         data_collator=default_data_collator,
-        compute_metrics=compute_metrics if training_args.do_eval and not is_torch_tpu_available() else None,
+        compute_metrics=compute_metrics
+        if training_args.do_eval and not is_torch_xla_available(check_is_tpu=True)
+        else None,
         preprocess_logits_for_metrics=(
-            preprocess_logits_for_metrics if training_args.do_eval and not is_torch_tpu_available() else None
+            preprocess_logits_for_metrics
+            if training_args.do_eval and not is_torch_xla_available(check_is_tpu=True)
+            else None
         ),
     )
 

examples/pytorch/language-modeling/run_fim_no_trainer.py

@@ -52,7 +52,7 @@ from transformers import (
     SchedulerType,
     default_data_collator,
     get_scheduler,
-    is_torch_tpu_available,
+    is_torch_xla_available,
 )
 from transformers.integrations import is_deepspeed_zero3_enabled
 from transformers.utils import check_min_version, send_example_telemetry
@@ -60,7 +60,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 logger = get_logger(__name__)
 
@@ -234,9 +234,7 @@ def parse_args():
         "--fim_pad_token",
         type=str,
         default="<fim_pad>",
-        help=(
-            "Fill-in-Middle Pad token. Used only when 'truncate_or_pad' is set to True." " Defaults to '<fim_pad>'."
-        ),
+        help=("Fill-in-Middle Pad token. Used only when 'truncate_or_pad' is set to True. Defaults to '<fim_pad>'."),
     )
     parser.add_argument(
         "--preprocessing_num_workers",
@@ -494,7 +492,7 @@ def main():
     if torch.cuda.is_availble():
         pad_factor = 8
 
-    elif is_torch_tpu_available():
+    elif is_torch_xla_available(check_is_tpu=True):
         pad_factor = 128
 
     # Add the new tokens to the tokenizer

examples/pytorch/language-modeling/run_mlm.py

@@ -54,7 +54,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")
 

examples/pytorch/language-modeling/run_mlm_no_trainer.py

@@ -57,7 +57,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 logger = get_logger(__name__)
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")

examples/pytorch/language-modeling/run_plm.py

@@ -47,7 +47,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=2.14.0", "To fix: pip install -r examples/pytorch/language-modeling/requirements.txt")
 

examples/pytorch/multiple-choice/run_swag.py

@@ -46,7 +46,7 @@ from transformers.utils import check_min_version, send_example_telemetry
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 logger = logging.getLogger(__name__)
 

examples/pytorch/multiple-choice/run_swag_no_trainer.py

@@ -54,7 +54,7 @@ from transformers.utils import check_min_version, send_example_telemetry
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 logger = get_logger(__name__)
 # You should update this to your particular problem to have better documentation of `model_type`

examples/pytorch/object-detection/README.md

@@ -56,7 +56,7 @@ python run_object_detection.py \
     --greater_is_better true \
     --load_best_model_at_end true \
     --logging_strategy epoch \
-    --evaluation_strategy epoch \
+    --eval_strategy epoch \
     --save_strategy epoch \
     --save_total_limit 2 \
     --push_to_hub true \

examples/pytorch/object-detection/run_object_detection.py

@@ -48,7 +48,7 @@ from transformers.utils.versions import require_version
 logger = logging.getLogger(__name__)
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=2.0.0", "To fix: pip install -r examples/pytorch/object-detection/requirements.txt")
 

examples/pytorch/object-detection/run_object_detection_no_trainer.py

@@ -51,7 +51,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 logging.basicConfig(level=logging.INFO)
 logger = get_logger(__name__)

examples/pytorch/question-answering/run_qa.py

@@ -50,7 +50,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt")
 

examples/pytorch/question-answering/run_qa_beam_search.py

@@ -48,7 +48,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt")
 

examples/pytorch/question-answering/run_qa_beam_search_no_trainer.py

@@ -55,7 +55,7 @@ from transformers.utils.versions import require_version
 
 
 # Will error if the minimal version of Transformers is not installed. Remove at your own risks.
-check_min_version("4.50.0.dev0")
+check_min_version("4.51.0.dev0")
 
 require_version("datasets>=1.8.0", "To fix: pip install -r examples/pytorch/question-answering/requirements.txt")