iHore/vLLM
8
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity
14,386 Commits 3 Branches 0 Tags
fa6a6be51978bd4b49ba0da17039e60f96dc5b13
Commit Graph
4625 Commits
Author SHA1 Message Date
chengchengpeiandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> 965525667b Onboard voyage-4-nano (#33720) 
Signed-off-by: Chengcheng Pei <chengchengpei@outlook.com>
Signed-off-by: chengchengpei <5881383+chengchengpei@users.noreply.github.com>
Co-authored-by: chengchengpei <5881383+chengchengpei@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
 2026-02-06 06:23:34 +00:00
Mingliang LilimingliangDarkLight1337
 		a32cb49b60 feat(frontend): early-fail tokenization guard for user requests (#31366) 
Signed-off-by: limingliang <limingliang@stepfun.com>
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Co-authored-by: limingliang <limingliang@stepfun.com>
Co-authored-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-02-05 19:38:02 -08:00
Xin Yang 79028d4388 [Perf] Disable clean_logits in deepgemm fp8_mqa_logits kernel (#33568) 2026-02-05 20:34:00 -05:00
emricksini-h 325ab6b0a8 [Feature] OTEL tracing during loading (#31162) 2026-02-05 16:59:28 -08:00
Hashem Hashemi d5c4800112 Adds padding and perf improvements to wvSplitK_fp8 (#33527) 
Signed-off-by: Hashem Hashemi <hashem.hashemi@amd.com>
 2026-02-05 22:16:02 +00:00
Cyrus Leung 116880a5a0 [Bugfix] Make MM batching more robust (#33817) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-02-05 20:40:58 +00:00
Harry Mellor 1887acca9e Fix tokenizer test for renamed attr on Transformers v5 (#33902) 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2026-02-05 19:16:20 +00:00
bnellnmandRobert Shaw a57c8228ff [Moe Refactor] Make Inplace Flag for FusedMoEModularKernel part of the constructor (#33375) 
Signed-off-by: Bill Nell <bnell@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
 2026-02-05 18:07:18 +00:00
Benjamin Chislett af3162d3aa [Spec Decode] Unified Parallel Drafting (#32887) 
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
 2026-02-05 12:37:18 -05:00
Aaron HaoandSumanthRH c1858b7ec8 [Feat][RL][1/2] Native Weight Syncing API: NCCL (#31943) 
Signed-off-by: ahao-anyscale <ahao@anyscale.com>
Signed-off-by: Aaron Hao <ahao@anyscale.com>
Co-authored-by: SumanthRH <sumanthrh99@gmail.com>
 2026-02-05 12:13:23 -05:00
Mario Hong 82914d2ae8 [Bugfix] Fix step3p5 parser when using mtp (#33690) 
Signed-off-by: mariohong <mariohong128@gmail.com>
 2026-02-05 16:04:04 +00:00
wang.yuqi 1c3a221d3b [Bugfix] Fix corner case of sparse embedding (#33886) 
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
 2026-02-05 02:51:22 -08:00
liranschourOr OzeriNicolò LucchesiNicolò Lucchesi
 		8322d4e47f Enable Cross layers KV cache layout at NIXL Connector V2 (#33339) 
Signed-off-by: Liran Schour <lirans@il.ibm.com>
Signed-off-by: liranschour <liranschour@users.noreply.github.com>
Co-authored-by: Or Ozeri <or@ozery.com>
Co-authored-by: Nicolò Lucchesi <nicolo.lucchesi@gmail.com>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
 2026-02-05 02:17:02 -08:00
Andreas KaratzasandMatthew Wong 3e472e81f9 [ROCm][Bugfix][CI] Fix hybrid models and their tests (Mamba/Jamba/Bamba) (#32710) 
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Signed-off-by: Matthew Wong <Matthew.Wong2@amd.com>
Co-authored-by: Matthew Wong <Matthew.Wong2@amd.com>
 2026-02-05 10:01:23 +00:00
Cyrus Leungandwang.yuqi 038914b7c8 [Refactor] Move task outside of PoolingParams.verify (#33796) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Co-authored-by: wang.yuqi <yuqi.wang@daocloud.io>
 2026-02-05 09:33:11 +00:00
Mark McLoughlin 2abd97592f [KV Connector][Metrics] Do not count local prefix cache hits in connector queries (#30522) 
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
 2026-02-05 09:57:27 +02:00
Andreas Karatzasandwang.yuqi 1f70313e59 [Bugfix] Fix ScoreMultiModalParam multi-document scoring returning single result (#33837) 
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
Co-authored-by: wang.yuqi <yuqi.wang@daocloud.io>
 2026-02-05 06:17:00 +00:00
rasmith c1395f72cd [CI][AMD][BugFix] Ensure VLLM_ROCM_USE_AITER is set so test_rocm_aiter_topk.py can run correctly (#33840) 
Signed-off-by: Randall Smith <Randall.Smith@amd.com>
 2026-02-05 05:05:48 +00:00
Nick Hill add9f1fbd9 [Minor] Include StreamingInput in inputs package (#33856) 
Signed-off-by: Nick Hill <nickhill123@gmail.com>
 2026-02-05 04:38:20 +00:00
Andreas Karatzas fb1270f1f8 [CI][Bugfix]: return McpCall for built-in MCP tools in non-streaming mode (#32762) 
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
 2026-02-05 11:14:06 +08:00
Luka Govedič 4d9513537d [CI][torch.compile] Reduce e2e fusion test time (#33293) 
Signed-off-by: Luka Govedič <lgovedic@redhat.com>
Signed-off-by: ProExpertProg <luka.govedic@gmail.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
 2026-02-04 19:09:03 -05:00
Ilya BoytsovCyrus Leungwang.yuqi
 		439afa4eea feat: Add ColBERT late interaction model support (#33686) 
Signed-off-by: Ilya Boytsov <ilyaboytsov1805@gmail.com>
Signed-off-by: Ilya Boytsov <boytsovpanamera@mail.ru>
Co-authored-by: Cyrus Leung <cyrus.tl.leung@gmail.com>
Co-authored-by: wang.yuqi <yuqi.wang@daocloud.io>
 2026-02-05 08:05:13 +08:00
Nick Hill fa4e0fb028 [Core] Don't schedule spec tokens with prefill chunks (#33652) 
Signed-off-by: Nick Hill <nickhill123@gmail.com>
 2026-02-04 23:40:22 +00:00
Richard Zou 9f14c9224d Revert "[torch.compile] Significantly speed up cold start times" (#33820) 
Signed-off-by: Richard Zou <zou3519@gmail.com>
 2026-02-04 21:59:59 +00:00
Simon Danielsson 4292c90a2a [Bugfix] Support RotaryEmbedding CustomOp for gpt-oss (#33800) 
Signed-off-by: simondanielsson <simon.danielsson99@hotmail.com>
 2026-02-04 20:17:41 +00:00
kourosh hakhamaneshi 2f6d17cb2f [rocm][ray] Fix: Unify Ray device visibility handling across CUDA and ROCm (#33308) 
Signed-off-by: Kourosh Hakhamaneshi <kourosh@anyscale.com>
 2026-02-04 10:09:14 -08:00
Isotr0py 192ad4648b [Bugfix] Fix interns1-pro initialization and PP (#33793) 
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
 2026-02-04 17:54:45 +00:00
Cyrus Leung 80f921ba4b [Bugfix] Fix normalize still being passed to PoolerConfig (#33794) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-02-04 23:56:02 +08:00
Or Ozeri 8e32690869 [KV Connector][BugFix] scheduler: Delay freeing blocks of aborted async loads (#32255) 
Fixes a not-yet-reported case where it was possible for blocks to be
freed by an abort before an async transfer completed, resulting
in corrupted KV data.

Signed-off-by: Or Ozeri <oro@il.ibm.com>
 2026-02-04 11:16:34 +00:00
zhanqiuhu 4403e3ed4c [Metrics] Add labeled prompt token metrics for P/D disaggregation (#33290) 
Add labeled Prometheus metrics to distinguish where prompt tokens come
from in P/D disaggregated deployments.

In P/D disaggregation, decode instances receive KV cache from prefill instances.
Currently, decode reports inflated prompt throughput because it counts all
prompt tokens as "computed", even though most were transferred.

This PR adds labeled metrics so users can understand actual compute work vs
transferred work:

vllm:prompt_tokens_by_source_total{source="local_compute"}        # Tokens prefilled locally
vllm:prompt_tokens_by_source_total{source="external_kv_transfer"} # Tokens received via KV transfer  
vllm:prompt_tokens_by_source_total{source="local_cache_hit"}      # Tokens from local prefix cache
vllm:prompt_tokens_cached_total                                    # Total cached (local + external, -1 when all 

Signed-off-by: Zhanqiu Hu <zh338@cornell.edu>
 2026-02-04 07:46:48 +00:00
Frank Wang 45f8fd6f97 [Feature] Enable TRITON_ATTN for Batch Invariance (#33688) 
Signed-off-by: frankwang28 <frank.wbb@hotmail.com>
 2026-02-04 13:27:34 +08:00
R3hankhan 4dffc5e044 [CPU] Split attention dispatch by head_dim alignment (#32161) 
Signed-off-by: Rehan Khan <Rehan.Khan7@ibm.com>
 2026-02-03 19:37:15 -08:00
Andrew XiaandAndrew Xia e1bf04b6c2 [1/N] Initial Implementation of Parser for ResponsesAPI (#32712) 
Signed-off-by: Andrew Xia <axia@fb.com>
Co-authored-by: Andrew Xia <axia@fb.com>
 2026-02-04 10:59:03 +08:00
Isotr0py 02080179a3 [Bugfix] Fix torchrun PP broadcast deadlock with async scheduling (#33701) 
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
 2026-02-04 02:17:37 +00:00
wang.yuqi 1b8fe6f7c4 [Frontend][4/n] Make pooling entrypoints request schema consensus | ScoreRequest (#33060) 
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
 2026-02-04 01:48:40 +00:00
Nick Hill 52ee21021a [BugFix][Spec Decoding] Fix negative accepted tokens metric crash (#33729) 
Signed-off-by: Nick Hill <nickhill123@gmail.com>
 2026-02-03 23:34:41 +00:00
Patrick von Platen 3f7662d650 [Voxtral Realtime] Change name (#33716) 
Signed-off-by: Patrick von Platen <patrick.v.platen@gmail.com>
 2026-02-03 13:03:28 -08:00
Harry Mellor 61e632aea1 Turn @config into a dataclass_transform (#31541) 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2026-02-03 17:40:59 +00:00
Richard Zou b1bb18de8d [torch.compile] Significantly speed up cold start times (#33641) 
Signed-off-by: Richard Zou <zou3519@gmail.com>
 2026-02-03 09:12:11 -08:00
shaharmor98 4bc913aeec Feat/add nemotron nano v3 tests (#33345) 2026-02-03 08:52:49 -05:00
zxyandIsotr0py a3acfa1071 [Models] Intern-S1-Pro (#33636) 
Signed-off-by: zxy <zhou0493@e.ntu.edu.sg>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
 2026-02-03 05:49:45 -08:00
Harry Mellor f6af34626d Fix offline test for Transformers v5 (#33682) 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2026-02-03 12:07:24 +00:00
Cyrus Leung 83449a5ff0 [Refactor] Clean up pooling serial utils (#33665) 
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
 2026-02-03 10:29:18 +00:00
Isotr0py 32e84fa1ff [CI/Build] Investigate torchrun distributed tests hanging issue (#33650) 
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
 2026-02-03 15:49:17 +08:00
杨朱 · KikiandClaude Opus 4.5 b95cc5014d [Misc] Remove deprecated VLLM_ALL2ALL_BACKEND environment variable (#33535) 
Signed-off-by: carlory <baofa.fan@daocloud.io>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
 2026-02-03 15:01:59 +08:00
Daniel MeschederandDaniel Mescheder 4c4b6f7a97 [Frontend] Add sampling parameters to Responses API (#32609) 
Signed-off-by: Daniel Mescheder <dmesch@amazon.com>
Co-authored-by: Daniel Mescheder <dmesch@amazon.com>
 2026-02-03 13:51:10 +08:00
Patrick von Platenandgemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> 5019c59dd2 [Voxtral Realtime] Introduce global log mel max (#33574) 
Signed-off-by: Patrick von Platen <patrick.v.platen@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
 2026-02-02 17:01:47 -05:00
Vasiliy Kuznetsov 0130223bd9 fix memory for online fp8 quantization with streaming weight load (#31914) 
Signed-off-by: vasiliy <vasiliy@fb.com>
 2026-02-02 14:17:42 -05:00
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
Harry Mellor 6141ebe0dd Remove incorrect tokenizer info test (#33565) 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
 2026-02-02 17:11:44 +00:00