iHore/vLLM
8
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity
14,386 Commits 3 Branches 0 Tags
fa6a6be51978bd4b49ba0da17039e60f96dc5b13
Commit Graph
2339 Commits
Author SHA1 Message Date
Kunshang Ji e10604480b [XPU][1/N] Deprecate ipex and switch to vllm-xpu-kernels for xpu platform (#33379) 
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
 2026-02-02 22:46:10 -08:00
yugong333 ffe1fc7a28 Reduce the kernel overhead when num of active loras is smaller than max loras. Multiple cuda graphs are captured for each num of active-loras. (#32005) 
Signed-off-by: Yu Gong <yu3.gong@gmail.com>
 2026-02-02 12:30:06 -05:00
Cyrus Leung d7e17aaacd [Refactor] Move profiling methods to MM budget (#33559) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-02-02 23:27:00 +08:00
Rabi Mishra 9eb58f8cf1 fix[ROCm]: Remove unconditional aiter import (#32902) 
Signed-off-by: rabi <ramishra@redhat.com>
 2026-02-02 22:10:02 +08:00
Cyrus Leung a502831d36 [Chore] Remove redundant input parsing methods (#33542) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-02-02 10:50:47 +00:00
Yifan Qiao a01ef3fa51 [Fix] prefix cache hit rate == 0 bug with gpt-oss style models (#33524) 
Signed-off-by: Yifan Qiao <yifanqiao@berkeley.edu>
 2026-02-02 01:59:58 +00:00
Nick Hill cf0a99f84d [ModelRunner V2] Support spec decode with structured outputs (#33374) 
Signed-off-by: Nick Hill <nickhill123@gmail.com>
 2026-02-02 00:19:59 +00:00
Nick Hill e535d90deb [ModelRunner V2] Misc minor simplifications and optimizations (#33467) 
Signed-off-by: Nick Hill <nickhill123@gmail.com>
 2026-02-01 22:17:14 +00:00
Komal Kumar Teru 0b225fb7b2 [Misc] skip target model mm emb in draft proposal step when draft is text-only (#33437) 
Signed-off-by: kkt-cohere <komal@cohere.com>
 2026-02-01 21:13:35 +00:00
Cyrus LeungandYunzhuLu 21997f45b1 [Redo] #33110 with threading limit (#33502) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: YunzhuLu <lucia.yunzhu@gmail.com>
 2026-02-01 09:18:11 +00:00
Cyrus Leung b6bb2842cf [Critical] Revert #33110 (#33500) 2026-01-31 21:06:42 -08:00
Cyrus Leung 79b6ec6aab [Bugfix] Fix inconsistent handling of cache reset (#33481) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-01-31 20:23:41 -08:00
Cyrus Leung a358e4dffe [Refactor] Make Renderer an abstract class (#33479) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-02-01 10:36:30 +08:00
smashyalts 1e86c802d4 Fix grammar (#33121) 
Signed-off-by: smashyalts <smashyalts@gmail.com>
 2026-01-31 09:59:34 -08:00
Cyrus Leung 88c3e114d8 [Refactor] Move MM data parsing outside processor (#33408) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-01-31 16:46:14 +00:00
YunzhuLuCyrus LeungRoger Wang
 		27cb2f678f [Bugfix] Early-reject requests with MM data longer than encode cache capacity (#33110) 
Signed-off-by: YunzhuLu <lucia.yunzhu@gmail.com>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
 2026-01-31 08:41:13 -08:00
jma99_2333andRoger Wang 22d9a056d5 Support clear mm and encoder cache (#33452) 
Signed-off-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Roger Wang <hey@rogerw.io>
 2026-01-31 15:22:25 +00:00
ℍ𝕠𝕝𝕝𝕠𝕨 𝕄𝕒𝕟 13b842f271 [BugFix][Router Replay] Capture Logical Experts with EPLB (#33013) 
Signed-off-by: Hollow Man <hollowman@opensuse.org>
 2026-01-31 10:12:17 -05:00
Cyrus Leung f0a1c8453a [Frontend] Use new Renderer for Completions and Tokenize API (#32863) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-01-31 04:51:15 -08:00
caozuoba 8980001c93 [perf] v1/spec_decode: skip softmax for all-greedy rejection sampling (#32852) 
Signed-off-by: hdj <1293066020@qq.com>
 2026-01-31 09:51:26 +00:00
jennyyyyzhen 527bcd14d4 [ROCM] Enable aiter attn backend for qwen3-next model (#32492) 
Signed-off-by: jennyyyyzhen <yzhen@hmc.edu>
 2026-01-31 17:03:57 +08:00
Francesco Fusco 5b55c0bea7 [Attention] Clarify comment explaining attn_logits +1 dimension (#33427) 
Signed-off-by: Francesco Fusco <ffu@zurich.ibm.com>
 2026-01-31 04:50:30 +00:00
Russell Bryant a2ef06e1b3 [Misc] offest -> offset in comments and variable names (#33444) 
Signed-off-by: Russell Bryant <rbryant@redhat.com>
 2026-01-30 20:19:22 -08:00
Lucas Wilkinsongemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>Michael Goin
 		0a3c71e7e5 [BugFix] Fix whisper FA2 + full cudagraphs (#33360) 
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
 2026-01-31 12:15:06 +08:00
Nick Hill 876a16f4fb [ModelRunner V2] Fix spec decoding + logprobs (#33391) 
Signed-off-by: Nick Hill <nickhill123@gmail.com>
 2026-01-31 03:33:26 +00:00
Matthew Bonanni aaa901ad55 [Attention] Move MLA forward from backend to layer (#33284) 
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
 2026-01-30 19:30:00 -08:00
Wentao Ye 010ec0c30e [Deprecation] Deprecate seed_everything and scatter_mm_placeholders in v0.15 (#33362) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-01-31 02:54:16 +00:00
Alberto Ferrer 64a40a7ab4 [Bugfix] Fix typo in read_offset variable name (#33426) 
Signed-off-by: Alberto Ferrer <albertof@barrahome.org>
 2026-01-31 01:26:15 +00:00
Kyle Sayers f857a03f6b [QeRL] Layerwise Reloading (#32133) 
Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>
 2026-01-30 08:50:05 -07:00
Patrick von PlatenandNick Hill 10152d2194 [Realtime API] Adds minimal realtime API based on websockets (#33187) 
Signed-off-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
 2026-01-30 18:41:29 +08:00
杨朱 · KikiandClaude Opus 4.5 1a7894dbdf [Misc] Replace Optional[X] with X | None syntax (#33332) 
Signed-off-by: carlory <baofa.fan@daocloud.io>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
 2026-01-30 01:56:59 -08:00
Cyrus Leung c87eac18f7 [Refactor] Move MM item count validation outside of processor (#33396) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-01-30 09:27:31 +00:00
Lucas Kabela 726d89720c [CI] Enable mypy import following for vllm/spec_decode (#33282) 
Signed-off-by: Lucas Kabela <lucaskabela@meta.com>
 2026-01-30 06:43:32 +00:00
Chendi.Xue 8c8ebeb941 [BUGFIX][XPU] fix memory check after XPU reuse GPU_worker (#33358) 
Signed-off-by: Chendi Xue <chendi.xue@intel.com>
 2026-01-29 09:56:30 -08:00
Cyrus Leung c6e7404cc5 [Multimodal] Simplify MM input definitions (#33331) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-01-29 13:32:04 +00:00
Kunshang Ji 8bb6271c77 [Intel GPU] refine xpu worker (#32894) 
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
 2026-01-29 12:26:52 +00:00
Harry Mellor fb946a7f89 Make mypy opt-out instead of opt-in (#33205) 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2026-01-29 09:12:26 +00:00
Didier Durandandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> 31b25f6516 [Doc]: fixing multiple typos in diverse files (#33256) 
Signed-off-by: Didier Durand <durand.didier@gmail.com>
Signed-off-by: Didier Durand <2927957+didier-durand@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
 2026-01-29 16:52:03 +08:00
Nick Hill 6bf3b46d78 [ModelRunner V2] Misc code simplification and cleanup (#33266) 
Signed-off-by: Nick Hill <nickhill123@gmail.com>
 2026-01-28 14:41:23 -08:00
Gregory Shtrasberg ab597c869a [Bugfix] Add missing encoder only guard for do_kv_cache_update (#33269) 
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
 2026-01-28 21:25:07 +00:00
Wentao YeandNick Hill 3e440786af [Feature] Fully support for async scheduling + PP, 30.8% E2E throughput improvement, 31.8% TPOT improvement (#32618) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Signed-off-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
 2026-01-28 20:30:32 +00:00
Gregory Shtrasberg 22ad649501 [ROCm] Enabling forward_includes_kv_cache on ROCm MHA backends (#33106) 
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
 2026-01-28 14:36:14 +08:00
22quinn a2b877df6c [Bugfix] Lazy import NgramProposer in GPU model runner (#32821) 
Signed-off-by: 22quinn <33176974+22quinn@users.noreply.github.com>
 2026-01-27 21:07:16 -08:00
Harry Mellor 2eb673a088 Add flake8-implicit-str-concat rules to Ruff (#33191) 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2026-01-28 04:56:10 +00:00
Woosuk KwonandNick Hill 46ec6d71c7 [Model Runner V2] Use a different stream for grammar bitmask h2d copy (#33059) 
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Signed-off-by: Woosuk Kwon <woosuk.kwon@berkeley.edu>
Co-authored-by: Nick Hill <nhill@redhat.com>
 2026-01-27 16:37:43 -08:00
Wentao Ye 3a6d5cbefd [Perf] Optimize dcp allocate tensor (#33102) 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
 2026-01-27 17:24:41 -05:00
Iris bd92089d33 feature: support eagle3 for HunyuanVL & Hunyuan (#33035) 
Signed-off-by: irisliu10 <601012173@qq.com>
Signed-off-by: Iris <38269816+irisliu10@users.noreply.github.com>
 2026-01-27 17:55:48 +00:00
Nick Hill 0cd259b2d8 [BugFix] Fix P/D with non-MoE DP (#33037) 
Signed-off-by: Nick Hill <nickhill123@gmail.com>
 2026-01-27 08:03:47 -08:00
omkhalilandOmar Mohamed Khalil 5ec44056f7 [Metrics][MFU] Fix UnembedMetrics FLOP overcounting for prefill (#33045) (#33045) 
Fix UnembedMetrics to correctly count FLOPs for the unembedding (LM head) layer.

The bug: UnembedMetrics used total_num_tokens() which counts all tokens in the
batch for projection flops, vocab projections are run on just the last token for the
autoregressive use case.

Co-authored-by: Omar Mohamed Khalil <omarkhalil@meta.com>
 2026-01-27 15:16:49 +00:00
Matthew Bonanni a608b4c6c2 [5/N][Attention] Finish eliminating vllm/attention folder (#32064) 
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
 2026-01-27 10:02:51 -05:00