|
|
a258ad8bcc
|
[Bugfix] fix qwen3 moe fp8 accuracy issue (#23031)
Signed-off-by: Jinzhen Lin <jinzhen.ljz@antgroup.com>
|
2025-08-16 17:41:23 -07:00 |
|
|
|
bf7f470b22
|
[V1] Logits processors extensibility (#19912)
Signed-off-by: Andrew Feldman <afeldman@redhat.com>
Signed-off-by: Andrew Feldman <afeld2012@gmail.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Signed-off-by: Nick Hill <nhill@redhat.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
Co-authored-by: Andrew Feldman <afeld2012@gmail.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2025-08-16 12:59:17 -07:00 |
|
|
|
4fc722eca4
|
[Kernel/Quant] Remove AQLM (#22943)
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
|
2025-08-16 19:38:21 +00:00 |
|
|
|
3253ae765e
|
[Flaky CI] Increase timeout tolerance for test_mp_crash_detection+test_default_mm_lora_chat_completions (#23028)
Signed-off-by: mgoin <mgoin64@gmail.com>
|
2025-08-16 18:33:08 +00:00 |
|
|
|
000cceca8c
|
[Bugfix gpt-oss] Fix float32 convert for flashinfer sink support (#23016)
Signed-off-by: mgoin <mgoin64@gmail.com>
|
2025-08-16 11:16:00 -07:00 |
|
|
|
68373d3126
|
[Frontend] Added support for HermesToolParser for models without special tokens (#16890)
Signed-off-by: minpeter <kali2005611@gmail.com>
|
2025-08-16 17:38:42 +00:00 |
|
|
|
52ce1420e9
|
Fix handling of max_num_batched_tokens for pooling tasks (#23004)
Signed-off-by: Max de Bayser <mbayser@br.ibm.com>
|
2025-08-16 17:36:30 +00:00 |
|
|
|
829bbd7882
|
[New Model]mBART model (#22883)
Signed-off-by: 汪志鹏 <wangzhipeng628@gmail.com>
|
2025-08-16 12:16:58 +00:00 |
|
|
|
4dff91c93d
|
[Refactor] Allow optional MultiModalKwargsItem in IPC (#23022)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
|
2025-08-16 11:30:49 +00:00 |
|
|
|
de9cb61763
|
Add docs for PrefixRepetitionDataset + enable usage with vllm bench throughput (#23012)
Signed-off-by: Seiji Eicher <seiji@anyscale.com>
Co-authored-by: Roger Wang <hey@rogerw.me>
|
2025-08-16 10:21:20 +00:00 |
|
|
|
2dbccce8a6
|
[CI][Bugfix] Skip Ovis2 generation test because of broken remote code (#22954)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
|
2025-08-16 09:44:19 +00:00 |
|
|
|
933f45334a
|
[Core] Make cudagraph check cuda platform only (#23005)
Signed-off-by: Chengji Yao <chengjiyao@gmail.com>
Signed-off-by: Chengji Yao <chengjiyao@google.com>
Co-authored-by: Chengji Yao <chengjiyao@gmail.com>
Co-authored-by: Li, Jiang <jiang1.li@intel.com>
|
2025-08-16 07:46:00 +00:00 |
|
|
|
cc826a202b
|
[Multimodal] Update Tensor schema test to cover arbitrary shape mm inputs (#22867)
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
|
2025-08-16 00:44:50 -07:00 |
|
|
|
6d3da472bc
|
[Misc] Add --save-dir option to benchmark_moe (#23020)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
|
2025-08-16 07:26:10 +00:00 |
|
|
|
78863f8c5c
|
[BugFix] Add support for loading prompt embeds tensors serialized on unavailable devices and sparse tensors (#22962)
Signed-off-by: Andrew Sansom <andrew@protopia.ai>
|
2025-08-16 06:25:10 +00:00 |
|
|
|
5157827cfc
|
[Build] Env var to disable sccache (#22968)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
|
2025-08-16 05:36:27 +00:00 |
|
|
|
7caec10e7b
|
[XPU]avoid circular import during XPU init (#23017)
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
|
2025-08-16 05:16:34 +00:00 |
|
|
|
1f83e7d849
|
[misc] nsys profile output kernel classifier and visualizer (#22971)
Signed-off-by: Grace Ho <grho@nvidia.com>
|
2025-08-16 02:52:51 +00:00 |
|
|
|
e4e37ded56
|
[V1] support min_tokens for detokener (#22014)
Signed-off-by: calvin chen <wen.chen@dynamia.ai>
Co-authored-by: Nick Hill <nhill@redhat.com>
|
2025-08-16 02:28:10 +00:00 |
|
|
|
f6b5040590
|
[Frontend] Avoid list copies in serving_chat.py (#22947)
Signed-off-by: Nick Hill <nhill@redhat.com>
|
2025-08-16 02:06:30 +00:00 |
|
|
|
fbd88728b3
|
[Bugfix] Fix DeepSeek MTP (#22934)
Signed-off-by: Benjamin Chislett <benjamin.chislett@centml.ai>
|
2025-08-16 01:25:06 +00:00 |
|
|
|
070da660c1
|
[Kernel] Simplify get_kv_cache_layout and cache use_trtllm_attention env-dependent bit (#22735)
Signed-off-by: NickLucche <nlucches@redhat.com>
|
2025-08-16 00:14:08 +00:00 |
|
|
|
ad0297d113
|
[Misc] Support passing multiple request ids at once to AsyncLLM.abort() (#22944)
Signed-off-by: Nick Hill <nhill@redhat.com>
|
2025-08-15 17:00:36 -07:00 |
|
|
|
236b864e4f
|
[BugFix] Make run_once thread-safe (#22978)
Signed-off-by: <wenji.yyc@alibaba-inc.com>
Signed-off-by: Yichen Yan <wenji.yyc@alibaba-inc.com>
|
2025-08-15 16:56:17 -07:00 |
|
|
|
3e2f7985a2
|
Support multiple attention groups for KV sharing (#22672)
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
|
2025-08-15 16:54:10 -07:00 |
|
|
|
c280066f9d
|
[v1] Move block_hashes from KVCacheManager to Request.block_hashes (#19728)
Signed-off-by: Or Ozeri <oro@il.ibm.com>
|
2025-08-15 16:52:52 -07:00 |
|
|
|
b9dc9d2607
|
[BugFix] Handle case where async utility call is cancelled (#22996)
Signed-off-by: Nick Hill <nhill@redhat.com>
Co-authored-by: Yinghai Lu <yinghai@thinkingmachines.ai>
|
2025-08-15 17:38:42 -06:00 |
|
|
|
1fc375dc05
|
[Structured Outputs] [Bug] Fix misalignment in apply_grammar_bitmask causing unintended masking and NaN logits (#22963)
Signed-off-by: rishitdholakia13 <rishit+github@cohere.com>
|
2025-08-15 23:25:05 +00:00 |
|
|
|
76144adf76
|
ci: Add CUDA + arm64 release builds (#21201)
Signed-off-by: Eli Uriegas <eliuriegas@meta.com>
|
2025-08-15 23:16:23 +00:00 |
|
|
|
f5d412bafb
|
[BugFix] Fix regression caused by mamba state dtype PR (#22998)
Signed-off-by: Thomas Parnell <tpa@zurich.ibm.com>
|
2025-08-15 22:55:26 +00:00 |
|
|
|
177e55e3bd
|
[Attention] FA3 Attention Sinks Perf Boost (#22478)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
|
2025-08-15 17:41:07 -04:00 |
|
|
|
1723ef1aae
|
minor: zero workspace buffer init for flashinfer trtllm-gen attn (#22603)
|
2025-08-15 21:38:10 +00:00 |
|
|
|
00d6cba0cf
|
Add PrefixRepetitionRandomDataset to vllm bench serve datasets (#20638)
Signed-off-by: Seiji Eicher <seiji@anyscale.com>
|
2025-08-15 14:09:23 -07:00 |
|
|
|
7f89ed248f
|
[Fix] enable swap_ab for pplx problem size computation (#22991)
Signed-off-by: Shixian Cui <shixian@amazon.com>
Co-authored-by: Shixian Cui <shixian@amazon.com>
|
2025-08-15 14:02:12 -07:00 |
|
|
|
8a87cd27d9
|
[CI] Speed up Whisper tests by reusing server (#22859)
Signed-off-by: mgoin <mgoin64@gmail.com>
|
2025-08-15 16:56:31 -04:00 |
|
|
|
a344a1a7da
|
Use regex in convert-results-json-to-markdown.py (#22989)
Signed-off-by: Michael Goin <mgoin64@gmail.com>
|
2025-08-15 20:54:20 +00:00 |
|
|
|
79899b63f6
|
[Bugfix] Added more env vars to hash (#22449)
Signed-off-by: Julien Lin <jullin@nvidia.com>
|
2025-08-15 20:08:37 +00:00 |
|
|
|
6e670778cd
|
[Core] direct indexing on self.block_table_np in compute_slot_mapping (#22940)
Signed-off-by: linzebing <linzebing1995@gmail.com>
|
2025-08-15 12:12:12 -07:00 |
|
|
|
df5afa82e5
|
[Log] Debug Once for Randomizing dummy data for DP Rank (#22860)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
|
2025-08-15 11:51:50 -07:00 |
|
|
|
6cd69f51bf
|
[Model] Granite-4 support loading quantized checkpoint (#22925)
Signed-off-by: Chih-Chieh-Yang <7364402+cyang49@users.noreply.github.com>
|
2025-08-15 18:47:56 +00:00 |
|
|
|
8ad7285ea2
|
[Kernels] Clean up FusedMoeMethodBase and modular kernel setup. Remove extra arguments from modular kernel methods. (#22035)
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
|
2025-08-15 14:46:00 -04:00 |
|
|
|
48b01fd4d4
|
[Structured Output] Make the output of structured output example more complete (#22481)
Signed-off-by: shen-shanshan <467638484@qq.com>
|
2025-08-15 18:29:25 +00:00 |
|
|
|
993d3d122b
|
[Benchmarks] Include image data when ShareGPT4V dataset is used. (#22955)
Signed-off-by: Chenheli Hua <huachenheli@outlook.com>
|
2025-08-15 18:23:06 +00:00 |
|
|
|
68af77e51c
|
[FIXBUG] Correctly Apply Grammar Bitmask in Mixed Batches (#22896)
Signed-off-by: JartX <sagformas@epdcenter.es>
|
2025-08-15 17:42:49 +00:00 |
|
|
|
6b04039a72
|
[BugFix] Skip the Q component for QKVParallelLinear in the case of QKVCrossParallelLinear since its width is 0 (#22369)
Signed-off-by: sstamenk <sstamenk@amd.com>
|
2025-08-15 17:17:31 +00:00 |
|
|
|
1c859a1387
|
[V0 Deprecation] Remove advance_step (#22969)
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
|
2025-08-15 08:22:31 -07:00 |
|
|
|
74f441f4b5
|
[Core] Allow full cudagraph with separate attention routines and orthogonal to compilation, add support for FA2 and FlashInfer (#20059)
Signed-off-by: fhl <2410591650@qq.com>
Signed-off-by: fhl2000 <63384265+fhl2000@users.noreply.github.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
|
2025-08-15 10:01:39 -04:00 |
|
|
|
a0632a3e03
|
[Frontend] Expose do_log_stats interval to env (#22905)
Signed-off-by: Csrayz <jover@cmbchina.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2025-08-15 13:00:20 +00:00 |
|
|
|
e8b40c7fa2
|
[CI] Remove duplicated docs build from buildkite (#22924)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2025-08-15 05:58:06 -07:00 |
|
|
|
48f4636927
|
[Misc] Ignore ep_kernels_workspace (#22807)
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
|
2025-08-15 05:58:03 -07:00 |
|
