d2f6daf6a7
[audio hash update] update the pinned audio hash ( #162892 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162892
Approved by: https://github.com/pytorchbot
2025-09-14 04:27:37 +00:00
e74b21d66a
[vllm hash update] update the pinned vllm hash ( #162891 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned vllm hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162891
Approved by: https://github.com/pytorchbot
2025-09-14 04:27:35 +00:00
f0fcf436c5
[audio hash update] update the pinned audio hash ( #162864 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162864
Approved by: https://github.com/pytorchbot
2025-09-13 04:17:21 +00:00
5663910472
[vllm hash update] update the pinned vllm hash ( #162751 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned vllm hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162751
Approved by: https://github.com/pytorchbot
2025-09-13 04:16:51 +00:00
66133b1ab7
Build vLLM aarch64 nightly wheels ( #162664 )
...
PyTorch has published its aarch64 nightly wheels for all CUDA version after https://github.com/pytorch/pytorch/pull/162364
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162664
Approved by: https://github.com/atalman
2025-09-13 03:43:55 +00:00
f7ea4975ab
update the baseline data for the operator benchmark ( #162693 )
...
According to the results of the last four operator benchmark runs, we found that five models achieved more than a 30% improvement compared to the baseline. Therefore, we will update the operator benchmark baseline data.
We use the average results from the four runs as the new baseline for the five models.
And add a pull request trigger for the operator benchmark workflow
Benchmarking Framework | Benchmarking Module Name | Case Name | tag | run_backward | baseline old | r1 | r2 | r3 | r4 | avg | speedup
-- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | --
PyTorch | add | add_M1_N1_K1_cpu | short | FALSE | 3.9497 | 2.57 | 2.54 | 2.38 | 2.31 | 2.45 | 1.61
PyTorch | functional.hardtanh | functional.hardtanh_dims(512 512)_contigFalse_inplaceFalse_dtypetorch.quint8 | short | FALSE | 67.118 | 50.02 | 49.80 | 46.78 | 48.94 | 48.88 | 1.37
PyTorch | relu6 | relu6_dims(512 512)_contigFalse_inplaceFalse_dtypetorch.quint8 | short | FALSE | 68.739 | 51.17 | 51.19 | 48.07 | 50.42 | 50.21 | 1.37
PyTorch | relu6 | relu6_dims(256 1024)_contigFalse_inplaceFalse_dtypetorch.quint8 | short | FALSE | 69.1875 | 51.97 | 52.77 | 50.00 | 51.24 | 51.50 | 1.34
PyTorch | functional.hardtanh | functional.hardtanh_dims(256 1024)_contigFalse_inplaceFalse_dtypetorch.quint8 | short | FALSE | 67.436 | 50.98 | 51.69 | 49.06 | 49.87 | 50.40 | 1.34
@chuanqi129 @huydhn @desertfire @jainapurva
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162693
Approved by: https://github.com/huydhn
2025-09-12 20:53:29 +00:00
e15686b40d
Remove actionable label from docathon label sync script ( #155713 )
...
Make sure we don't propagate actionable label in docathon sync label script.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/155713
Approved by: https://github.com/clee2000
2025-09-12 15:36:50 +00:00
da954f10d6
Bump protobuf from 5.29.4 to 5.29.5 in /.github/requirements ( #160844 )
...
Bumps [protobuf](https://github.com/protocolbuffers/protobuf ) from 5.29.4 to 5.29.5.
<details>
<summary>Commits</summary>
<ul>
<li><a href="f5de0a049585637662f7https://redirect.github.com/protocolbuffers/protobuf/issues/21858 ">#21858</a> from shaod2/py-cp-29</li>
<li><a href="05ba1a81041ef3f01c4669cca9b7f5https://redirect.github.com/protocolbuffers/protobuf/issues/21612 ">#21612</a>)</li>
<li><a href="21fdb7acdbhttps://redirect.github.com/protocolbuffers/protobuf/issues/20446 ">#20446</a>) (<a href="https://redirect.github.com/protocolbuffers/protobuf/issues/20904 ">#20904</a>)</li>
<li><a href="03c50e3874https://redirect.github.com/protocolbuffers/protobuf/issues/20853 ">#20853</a>)</li>
<li><a href="128f0aafd9https://redirect.github.com/protocolbuffers/protobuf/issues/20767 ">#20767</a>)</li>
<li><a href="bdd49bb141https://redirect.github.com/protocolbuffers/protobuf/issues/20755 ">#20755</a> from protocolbuffers/29.x-202503192110</li>
<li><a href="c65946848fhttps://github.com/protocolbuffers/protobuf/compare/v5.29.4...v5.29.5 ">compare view</a></li>
</ul>
</details>
<br />
[](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores )
Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`.
[//]: # (dependabot-automerge-start)
[//]: # (dependabot-automerge-end)
---
<details>
<summary>Dependabot commands and options</summary>
<br />
You can trigger Dependabot actions by commenting on this PR:
- `@dependabot rebase` will rebase this PR
- `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it
- `@dependabot merge` will merge this PR after your CI passes on it
- `@dependabot squash and merge` will squash and merge this PR after your CI passes on it
- `@dependabot cancel merge` will cancel a previously requested merge and block automerging
- `@dependabot reopen` will reopen this PR if it is closed
- `@dependabot close` will close this PR and stop Dependabot recreating it. You can achieve the same result by closing it manually
- `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency
- `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
- `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)
You can disable automated security fix PRs for this repo from the [Security Alerts page](https://github.com/pytorch/pytorch/network/alerts ).
</details>
Pull Request resolved: https://github.com/pytorch/pytorch/pull/160844
Approved by: https://github.com/msaroufim
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2025-09-12 04:23:03 +00:00
d959eb02cb
[audio hash update] update the pinned audio hash ( #162752 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162752
Approved by: https://github.com/pytorchbot
2025-09-12 04:18:54 +00:00
62f044e260
Bump setuptools from 72.1.0 to 78.1.1 in /.github/requirements ( #162701 )
...
Bumps [setuptools](https://github.com/pypa/setuptools ) from 72.1.0 to 78.1.1.
- [Release notes](https://github.com/pypa/setuptools/releases )
- [Changelog](https://github.com/pypa/setuptools/blob/main/NEWS.rst )
- [Commits](https://github.com/pypa/setuptools/compare/v72.1.0...v78.1.1 )
---
updated-dependencies:
- dependency-name: setuptools
dependency-version: 78.1.1
dependency-type: direct:production
...
Signed-off-by: dependabot[bot] <support@github.com >
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2025-09-11 21:03:27 -07:00
e8eeb06034
Move inductor jobs 3.9->3.10 ( #162323 )
...
Related to: https://github.com/pytorch/pytorch/issues/161167
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162323
Approved by: https://github.com/huydhn , https://github.com/Skylion007
Co-authored-by: Huy Do <huydhn@gmail.com >
2025-09-12 03:43:06 +00:00
3cd734584d
bring back the old vllm's use_existing_torch.py ( #162747 )
...
vllm's pr will override our dependencies for torch.
quick fix to add the use_existing_torch.py. syncing with vllm now regarding the uv approach they have
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162747
Approved by: https://github.com/huydhn
2025-09-12 03:41:39 +00:00
222ec8d28e
Revert "AMD CPU CI - Add freezing + fix label trigger ( #162176 )"
...
This reverts commit 9cac1b92595ec7836101d51dbe1415081042c7a0.
Reverted https://github.com/pytorch/pytorch/pull/162176 on behalf of https://github.com/huydhn due to Sorry for reverting this but hardcoding the input online 122 does not make sense ([comment](https://github.com/pytorch/pytorch/pull/162176#issuecomment-3283532452 ))
2025-09-12 03:39:13 +00:00
84d8ec73f1
[CD] Build Mac wheels using setup-python action ( #162136 )
...
Biggest difference between both conda and homebrew CPython builds and one from python.org, is that later are universal binaries and they are always trying to build universal extension...
Workaround lots of universal binary build attempts by explicitly specifying both `_PYTHON_PLATFORM` and `--plat-name` as well as `ARCH_FLAGS`
Suppressed actionlint warning on use of `freethreaded` flag which is document in https://github.com/actions/setup-python/tree/v5
TODO: Remove lots of temporary workarounds when `3.14` is out in October 2025
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162136
Approved by: https://github.com/atalman , https://github.com/huydhn
ghstack dependencies: #162297 , #162265
2025-09-12 00:16:31 +00:00
9cac1b9259
AMD CPU CI - Add freezing + fix label trigger ( #162176 )
...
Added the following changes:
1. Added freezing by default for AMD CPU based CI
2. Fixed issue with label based CI triggers
Addresses code review comment in https://github.com/pytorch/pytorch/pull/161155
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162176
Approved by: https://github.com/malfet , https://github.com/jeffdaily
2025-09-11 18:41:29 +00:00
2f53395943
[ez][CI] Fix docs push in nightly workflow ( #162657 )
...
HUD metrics page says docs push hasn't happened in 21 days
<img width="293" height="142" alt="image" src="https://github.com/user-attachments/assets/f930aab8-0503-4bf2-b962-8c375dec6b78 " />
I guess main branch docs just haven't been updated? Did anyone notice? Do we care?
Either way I think this should fix it
Likely started after https://github.com/pytorch/pytorch/pull/161182
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162657
Approved by: https://github.com/huydhn
2025-09-11 16:45:41 +00:00
c3f30eca9e
Remove tests-to-include from rocm-mi300 workflow ( #162721 )
...
Accidentally introduced by https://github.com/pytorch/pytorch/pull/162288 (was meant to be a temporary change)
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162721
Approved by: https://github.com/jeffdaily
2025-09-11 14:36:07 +00:00
23170dfebc
Revert "Move inductor jobs 3.9->3.10 ( #162323 )"
...
This reverts commit 0663bdb12383b9717af49d58aed9d88de0dd0ecc.
Reverted https://github.com/pytorch/pytorch/pull/162323 on behalf of https://github.com/huydhn due to Not sure what had happened, but some inductor unit tests start failing after this lands ([comment](https://github.com/pytorch/pytorch/pull/162323#issuecomment-3278125192 ))
2025-09-11 05:57:13 +00:00
07d2531672
[vllm hash update] update the pinned vllm hash ( #162551 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned vllm hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162551
Approved by: https://github.com/pytorchbot
2025-09-11 04:56:04 +00:00
bb1d53bc47
[CD] CUDA 13 specific followup changes ( #162455 )
...
Follow up for CUDA 13 bring up https://github.com/pytorch/pytorch/issues/159779
sm50-70 should not be added to sbsa build arch list, as previous archs had no support for arm.
remove platform_machine from PYTORCH_EXTRA_INSTALL_REQUIREMENTS
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162455
Approved by: https://github.com/atalman
2025-09-11 00:03:47 +00:00
0663bdb123
Move inductor jobs 3.9->3.10 ( #162323 )
...
Related to: https://github.com/pytorch/pytorch/issues/161167
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162323
Approved by: https://github.com/huydhn , https://github.com/Skylion007
2025-09-10 20:58:41 +00:00
80d4da893c
Revert "Put torchao (0.13.0) back to benchmark workflow ( #162227 )"
...
This reverts commit 00985970e312c3c5e674e8e14d39fe77c226600e.
Reverted https://github.com/pytorch/pytorch/pull/162227 on behalf of https://github.com/huydhn due to Crashing some inductor jobs in trunk ([comment](https://github.com/pytorch/pytorch/pull/162227#issuecomment-3276355034 ))
2025-09-10 20:11:37 +00:00
ab0694f1c6
[ROCm][Inductor][CK backend] Install rocm-composable-kernel python package on ROCm Linux CI docker images ( #162288 )
...
Reopened from #158747 which got reverted since without setuptools-scm in pytorch index URL the wheel cannot be built
We reconsider the original PR idea of introducing CK as a pytorch dependency on ROCm Linux and install the CK python package in CI only -- since (1) rocm-composable-kernel depends on setuptools-scm which depends on tomli and the existing index URLs need to be modified to host the new packages and (2) there also is a packaging [bug](https://github.com/pypa/setuptools/issues/3269#issuecomment-1254507377 ) in Ubuntu 22.04 which prevents correct dynamic version calculation with default system pip.
Extras:
-> this PR reconsiders how TORCHINDUCTOR_CK_DIR env variable is used; previously, this var was used to point to rocm-composable-kernel package installation path on the filesystem; now, the path is inferred by trying to import ck4inductor
-> the tests are updated to reflect this change
-> since in CI clang points to a bash script which invokes sccache, we cannot patch PATH to not contain sccache, this logic is removed from the testing code
-> scaled_mm test crashes during the benchmarking when the benchmarking happens in the main process, and times out benchmarking when it happens in a subprocess, on gfx942, so it is disabled
TBD: roll back rocm-mi300 workflow before merging
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162288
Approved by: https://github.com/jeffdaily
2025-09-10 19:33:40 +00:00
11acfed3ce
[audio hash update] update the pinned audio hash ( #162552 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162552
Approved by: https://github.com/pytorchbot
2025-09-10 04:24:39 +00:00
e64965300a
Repackage vLLM nightlies ( #162371 )
...
I suspected that I would need to repack vLLM wheels from https://github.com/pytorch/pytorch/pull/162000 because I renamed the wheel, and it turns out to be true. The error is as follows:
```
$ uv pip install --pre xformers --index-url https://download.pytorch.org/whl/nightly/cu129
Using Python 3.12.11+meta environment at: venv/py3.12
Resolved 28 packages in 759ms
error: Failed to install: xformers-0.0.33.dev20250901+cu129-cp39-abi3-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (xformers==0.0.33.dev20250901+cu129)
Caused by: Wheel version does not match filename: 0.0.33+5d4b92a5.d20250907 != 0.0.33.dev20250901+cu129
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162371
Approved by: https://github.com/atalman
2025-09-10 04:02:34 +00:00
00985970e3
Put torchao (0.13.0) back to benchmark workflow ( #162227 )
...
0.13.0 was released on Sep 3rd https://pypi.org/project/torchao/#history , which should have fixed the crashing issue on transformers now
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162227
Approved by: https://github.com/malfet
2025-09-10 03:56:25 +00:00
8922bbcaab
Use same NVSHMEM version across CUDA builds ( #162206 )
...
#161321 bumped NVSHMEM version to 3.3.24 for CUDA 13, leaving CUDA 12 with 3.3.20.
This PR bumps the NVSHMEM version to 3.3.24 for CUDA 12 as well.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162206
Approved by: https://github.com/tinglvv , https://github.com/Skylion007
2025-09-09 20:59:50 +00:00
be3b8d2ec9
[ROCm][CI] update fbgemm nightly benchmark hash ( #162385 )
...
fbgemm_gpu was failing to clone due to missing submodule commit.
```
+ pushd fbgemm/fbgemm_gpu
~/pytorch/fbgemm/fbgemm_gpu ~/pytorch
+ git checkout 7f1de94a4c2d14f59ad4ca84538c36084ea6b2c8 --recurse-submodules
fatal: failed to unpack tree object b1281b8b08d973a7064f864f47eeb30f3e2596e9
error: Submodule 'external/composable_kernel' could not be updated.
error: Cannot update submodule:
external/composable_kernel
```
Log File
[inductor-periodic · pytorch/pytorch@5babb4d ](https://github.com/pytorch/pytorch/actions/runs/17536630806/job/49802458834 )
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162385
Approved by: https://github.com/jeffdaily
Co-authored-by: Jeff Daily <jeff.daily@amd.com >
2025-09-09 15:44:39 +00:00
5ccf3ca3ec
Revert "Use same NVSHMEM version across CUDA builds ( #162206 )"
...
This reverts commit 0d9c95cd7ee299e2e8c09df26d395be8775b506b.
Reverted https://github.com/pytorch/pytorch/pull/162206 on behalf of https://github.com/malfet due to Broke lint, see 4dd73e659a/1https://github.com/pytorch/pytorch/pull/162206#issuecomment-3271040521 ))
2025-09-09 14:40:45 +00:00
0d9c95cd7e
Use same NVSHMEM version across CUDA builds ( #162206 )
...
#161321 bumped NVSHMEM version to 3.3.24 for CUDA 13, leaving CUDA 12 with 3.3.20.
This PR bumps the NVSHMEM version to 3.3.24 for CUDA 12 as well.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162206
Approved by: https://github.com/tinglvv , https://github.com/Skylion007
2025-09-09 08:52:27 +00:00
4840a1a591
Run vLLM tests on all trunk commits before 2.9 branch cut ( #161797 )
...
This makes it easier to bisect issue now given that we don't have lots of time.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161797
Approved by: https://github.com/yangw-dev
2025-09-09 05:56:41 +00:00
d49205fe1f
Add more tests for vllm and clean out the old vllm test ( #162292 )
...
Test failure coverage from pytorch 2.8 release issues
[internal access only](https://docs.google.com/document/d/1zvK1eUAHubHGGHg9jKxd-QlP89fzgfqOBvE2m9mUs90/edit?tab=t.0
)
See coverage mapping
| Given test / pattern | Suite ID (from config) |
|---|---|
| pytest -v -s basic_correctness/test_cumem.py | vllm_basic_correctness_test |
| pytest -v -s entrypoints/openai/test_sleep.py | vllm_entrypoints_test |
| pytest -v -s entrypoints/openai/test_translation_validation.py::test_long_audio_request | vllm_entrypoints_test |
| pytest -v -s lora/test_quant_model.py | vllm_lora_28_failure_test |
| pytest -v -s -x tests/lora/test_llama_tp.py | vllm_lora_tp_test_distributed |
| pytest -v -s distributed/test_sequence_parallel.py -k test_tp_sp_generation |vllm_distributed_test_28_failure_test |
| pytest -v -s distributed/test_sequence_parallel.py::test_tp_sp_generation[...] | vllm_distributed_test_28_failure_test |
| pytest models/language/generation/test_mistral.py::test_models[...] | vllm_languagde_model_test_extended_generation_28_failure_test |
| pytest models/multimodal/pooling/test_jinavl_reranker.py::test_model_text_image[...] | vllm_multi_model_test_28_failure_test |
| tests/lora/test_qwen2vl.py::test_qwen2vl_lora | vllm_lora_test |
| tests/lora/test_qwen2vl.py::test_qwen25vl_lora | vllm_lora_test |
| tests/lora/test_qwen2vl.py::test_qwen2vl_lora_beam_search | vllm_lora_test |
| tests/lora/test_phi.py::test_phi2_lora | DIDN'T FIND IT IT IN VLLM |
| models/multimodal/generation/test_voxtral.py::test_models_with_multiple_audios[5-128-half] | vllm_multi_model_test_28_failure_test |
| models/test_initialization.py::test_can_initialize[VoxtralForConditionalGeneration] | vllm_basic_models_test |
| pytest -v -s -x lora/test_chatglm3_tp.py -k test_chatglm3_lora_tp4_fully_sharded_loras | vllm_lora_tp_test_distributed |
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162292
Approved by: https://github.com/atalman , https://github.com/huydhn
2025-09-09 05:53:46 +00:00
7ad40de60e
[audio hash update] update the pinned audio hash ( #162437 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162437
Approved by: https://github.com/pytorchbot
2025-09-09 04:41:34 +00:00
607327beae
[vllm hash update] update the pinned vllm hash ( #162356 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned vllm hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162356
Approved by: https://github.com/pytorchbot
2025-09-09 04:40:25 +00:00
dd44faa9d9
Revert "Modify ROCm MI2xx-based workflows to run on cron schedule ( #162103 )"
...
This reverts commit 0af70e2353e1dcda83175fd4834ecb7b63e009e0.
Reverted https://github.com/pytorch/pytorch/pull/162103 on behalf of https://github.com/jithunnair-amd due to Cirrascale network outage resolved. Reverting back to running per commit to aid in triage and CI health ([comment](https://github.com/pytorch/pytorch/pull/162103#issuecomment-3267977825 ))
2025-09-08 20:53:05 +00:00
9c991b63ff
[CD] [aarch64] Add CUDA 12.6 and 12.8 to build matrix, remove 12.9 build ( #162364 )
...
https://github.com/pytorch/pytorch/issues/159779
Add the full CUDA support matrix to sbsa build (12.6, 12.8)
Same arch support as x86 build
Remove 12.9 sbsa build
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162364
Approved by: https://github.com/atalman
2025-09-08 20:00:25 +00:00
314d47a210
[audio hash update] update the pinned audio hash ( #162315 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162315
Approved by: https://github.com/pytorchbot
2025-09-08 18:26:33 +00:00
53297f6ad0
Revert "[audio hash update] update the pinned audio hash ( #162315 )"
...
This reverts commit c9ac8c25ef9ad020542898ab569910a9d0cd1f7e.
Reverted https://github.com/pytorch/pytorch/pull/162315 on behalf of https://github.com/jeanschmidt due to Reverting in order to see if this introduced the failure https://github.com/pytorch/pytorch/actions/runs/17539536914/job/49810513700 ([comment](https://github.com/pytorch/pytorch/pull/162315#issuecomment-3266932718 ))
2025-09-08 15:52:30 +00:00
3f5993316e
[upstream triton] update triton pin to triton 3.5 ( #162278 )
...
Update PyTorch to the latest Triton release candidate branch (release/3.5.x in triton-lang/triton)
Notably:
* this does *not* include the version number bump from 3.4 -> 3.5 (we'll do that in a follow-up PR)
* sam_fast is still failing, so we've disabled it temporarily https://github.com/pytorch/pytorch/issues/162282 and we are committed to fixing it, ideally before the branch cut but possibly as a cherry-pick into the release branch.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162278
Approved by: https://github.com/atalman
ghstack dependencies: #162244 , #162309
2025-09-08 14:29:24 +00:00
32911ff541
[xla hash update] update the pinned xla hash ( #162372 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned xla hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162372
Approved by: https://github.com/pytorchbot
2025-09-08 11:31:16 +00:00
c9ac8c25ef
[audio hash update] update the pinned audio hash ( #162315 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162315
Approved by: https://github.com/pytorchbot
2025-09-08 04:17:23 +00:00
93fb23d6fa
Build vLLM nightly wheels ( #162000 )
...
This uses the same approach as building triton wheel where we publish a nightly wheel for vLLM whenever its pinned commit is updated. The key change is to use `pytorch/manylinux2_28-builder` as the base image to build vLLM, so there are a couple of changes on the vLLM Dockerfile used by lumen_cli
1. `pytorch/manylinux2_28-builder` is RedHat instead of Debian-based, so no apt-get
2. Fix a bug in `.github/actions/build-external-packages/action.yml` where `CUDA_VERSION` is not set correctly, preventing CUDA 12.9 build
3. Fix a bug in `.github/actions/build-external-packages/action.yml` where `TORCH_WHEELS_PATH` is not set correctly and always defaulted to `dist`
4. In vLLM Dockerfile, use the correct index for the selected CUDA version, i.e. https://download.pytorch.org/whl/nightly/cu12[89] for CUDA 12.[89]
5. Install torch, vision, audio in one command. Unlike the CI image `pytorch-linux-jammy-cuda12.8-cudnn9-py3.12-gcc11-vllm`, `pytorch/manylinux2_28-builder` doesn't have any torch dependencies preinstalled
6. Bump xformers version to 0.0.32.post2 now that PyTorch 2.8.0 has been landed on vLLM
We need to prepare 3 wheels for vLLM, xformers, and flashinfer-python. And I rename them in the same convention as PyTorch nightlies `MAJOR.MINOR.PATCH.devYYYYMMDD` so that vLLM nightlies will work with torch nightlies on the same date.
### Usage
* Install latest nightlies
```
pip install --pre torch torchvision torchaudio vllm xformers flashinfer_python \
--index-url https://download.pytorch.org/whl/nightly/cu129
```
* Install a specific version
```
pip install --pre torch==2.9.0.dev20250903 torchvision torchaudio \
vllm==1.0.0.dev20250903 \
xformers=0.0.33.dev20250903 \
flashinfer_python=0.2.14.dev20250903 \
--index-url https://download.pytorch.org/whl/nightly/cu129
```
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162000
Approved by: https://github.com/atalman
2025-09-07 06:09:17 +00:00
fea20775ad
[vllm hash update] update the pinned vllm hash ( #162314 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned vllm hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162314
Approved by: https://github.com/pytorchbot
2025-09-07 04:29:23 +00:00
145a3a7bda
[CUDA 13][cuDNN] Bump CUDA 13 to cuDNN 9.13.0 ( #162268 )
...
Fixes some `d_qk` != `d_v` cases on Hopper that are broken by cuDNN 9.11-9.12
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162268
Approved by: https://github.com/drisspg , https://github.com/Skylion007
2025-09-06 01:59:03 +00:00
8d50355d97
[CD][EZ] Update libtorch python version to 3.10 ( #162297 )
...
Not sure why it was at 3.9
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162297
Approved by: https://github.com/clee2000 , https://github.com/atalman
2025-09-05 22:46:36 +00:00
261a84a176
[CD][BE] Remove unnecessary checks for XCode version ( #162263 )
...
None of them have worked for a while, PyTorch for Mac is build with
XCode-15.4
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162263
Approved by: https://github.com/clee2000 , https://github.com/Skylion007 , https://github.com/ZainRizvi
2025-09-05 17:02:36 +00:00
bffc7dd1f3
[CD] Add cuda 13.0 libtorch builds, remove CUDA 12.9 builds ( #161916 )
...
Related to https://github.com/pytorch/pytorch/issues/159779
Adding CUDA 13.0 libtorch builds, followup after https://github.com/pytorch/pytorch/pull/160956
Removing CUDA 12.9 builds, See https://github.com/pytorch/pytorch/issues/159980
Pull Request resolved: https://github.com/pytorch/pytorch/pull/161916
Approved by: https://github.com/jeanschmidt , https://github.com/Skylion007
Co-authored-by: Ting Lu <tingl@nvidia.com >
2025-09-05 07:47:54 +00:00
494878a11b
[audio hash update] update the pinned audio hash ( #162114 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned audio hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162114
Approved by: https://github.com/pytorchbot
2025-09-05 04:32:16 +00:00
3bbc2e3e4f
[vllm hash update] update the pinned vllm hash ( #162226 )
...
This PR is auto-generated nightly by [this action](https://github.com/pytorch/pytorch/blob/main/.github/workflows/nightly.yml ).
Update the pinned vllm hash.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162226
Approved by: https://github.com/pytorchbot
2025-09-05 04:32:08 +00:00
6b8b3ac440
Revert "[ROCm] Use MI325 (gfx942) runners for binary smoke testing ( #162044 )"
...
This reverts commit cd529b686d54bbaa443f5b310140de48422d96c7.
Reverted https://github.com/pytorch/pytorch/pull/162044 on behalf of https://github.com/jeffdaily due to mi200 backlog is purged, and mi300 runners are failing in GHA download ([comment](https://github.com/pytorch/pytorch/pull/162044#issuecomment-3254427869 ))
2025-09-04 16:06:30 +00:00
