Commit Graph
419 Commits
Author SHA1 Message Date
Taksh Kothari 12b005eefa [Bugfix] Don't split a multimodal item that a prefill chunk starts exactly on (#52412)
Signed-off-by: Taksh <takshkothari09@gmail.com>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
2026-10-13 15:32:38 +00:00
Jiangyun ZhuClaude Opus 5.5mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2cc2e68d1c [Core][Mamba] Remove leftovers of mamba_cache_mode="all" (#60043)
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
Signed-off-by: Jiangyun Zhu <zhujiangyun@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-10-13 00:58:59 -07:00
df6c7ad031 [Core][Refactor] Move KV cache planning logic to kv_cache_planning.py (#61183)
Signed-off-by: Dakai An <dakaian108@gmail.com>
Signed-off-by: Yifan Qiao <yifanqiao@inferact.ai>
Co-authored-by: Yifan Qiao <yifanqiao@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-13 03:05:39 +00:00
ztysthandOpenAI Codex 1bdd3fa2d5 [Bugfix][Core] Check multimodal chunk boundaries before encoder cache reuse (#59721)
Signed-off-by: ztysth <rnk.liulw@gmail.com>
Signed-off-by: ztysth <166926442+ztysth@users.noreply.github.com>
Co-authored-by: OpenAI Codex <noreply@openai.com>
2026-10-12 19:25:35 +00:00
Woosuk KwonandClaude Opus 5.5 e1a5086256 [Attention] Remove cascade attention (#61126)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-12 10:41:07 -07:00
Nick Hill c059bee812 [MRV2] Switch CI tests still running on MRV1 unnecessarily (#61014)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
2026-10-11 09:57:07 -07:00
Matthew BonanniandClaude Opus 5.5 d5f356dae0 [HiSparse] Own GPU residency instead of editing core block-table rows (#60647)
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-09 21:25:44 -07:00
02e55d2af3 [KV Cache] GLM-5.3-Flash: use the generic packed KV layout; CircularBufferSpec tail with speculative ring slots (#57169)
Signed-off-by: Yifan Qiao <yifanqiao@inferact.ai>
Signed-off-by: Dakai An <dakaian108@gmail.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Yifan Qiao <yifanqiao@inferact.ai>
Co-authored-by: OpenAI Codex <noreply@openai.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
2026-10-09 15:48:37 -04:00
Wei ZhaoWei ZhaoCodexzjy0516Claude Opus 5.5mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
d8fa463e2b [Bugfix][Prefix caching] Fix inconsistent hybrid Mamba prefix cache boundaries (#60533)
Signed-off-by: Wei Zhao <weizha@aws-cmh-slurm-1-login-01.cm.cluster>
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
Co-authored-by: Wei Zhao <weizha@aws-cmh-slurm-1-login-01.cm.cluster>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: zjy0516 <riverclouds.zhu@qq.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-10-08 18:33:48 +00:00
Nicolò Lucchesi e161a0798c [Perf][Core] Reduce free KV cache queue overhead (#60269)
Signed-off-by: NickLucche <nicolo.lucchesi@mistral.ai>
2026-10-08 10:17:11 +00:00
7816afdb25 [Bugfix][Core] Drain async KV loads in pause(mode="wait"), and pause the drain test only once a request is in flight (#60572)
Signed-off-by: aoshen02 <aoshen@inferact.ai>
Signed-off-by: Stefan Koncarevic <Stefan.Koncarevic@amd.com>
Co-authored-by: aoshen02 <aoshen@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-08 09:33:41 +00:00
ef63c23d35 [Bugfix][DSv4.1] Skip SWA bounded replay for KV loads that carry the window (#59197)
Signed-off-by: Yifan Qiao <yifanqiao@inferact.ai>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-07 15:28:42 -07:00
e6fc81bc78 [Bugfix][Scheduler] Let pooling chunked prefill use full context (#48039)
Signed-off-by: Tommy <tran.tommy@hotmail.com>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
2026-10-07 07:32:00 +00:00
e37e51dd24 [Metrics] Expose cached prompt tokens by cache tier (#56318)
Signed-off-by: Cam Quilici <cjquilici@gmail.com>
Signed-off-by: Cam Quilici <cameron@semianalysis.com>
Co-authored-by: Cam Quilici <cameron@semianalysis.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Yifan Qiao <yifanqiao@inferact.ai>
2026-10-06 14:52:05 -04:00
6fa67afed9 [Bugfix][Core] Retire stale Mamba spec blocks in prefill checkpoint steps (#59759)
Signed-off-by: Simon Veitner <sveitner@redhat.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
2026-10-06 11:01:19 +08:00
2e3154aa18 [HiSparse] Log steady-state max concurrency and expose host-tier utilization gauges (#58949)
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
2026-10-05 14:20:32 -07:00
bzsuniandcjackal 528772a4bf [Bugfix][Core] Retain encoder cache references for repeated multimodal inputs (#59942)
Signed-off-by: bzsuni <bingzhe.sun@daocloud.io>
Signed-off-by: cjackal <44624812+cjackal@users.noreply.github.com>
Co-authored-by: cjackal <44624812+cjackal@users.noreply.github.com>
2026-10-05 15:52:46 +00:00
Nicolò Lucchesi 30d4032363 [Model][GLM-5.3-Flash] Support DCP for the kpool sparse indexer (#59211)
Signed-off-by: NickLucche <nicolo.lucchesi@mistral.ai>
2026-10-05 16:44:14 +02:00
ca0df1f122 [Bugfix][Core] Exempt exactly the blocks an async KV load writes from zeroing (#59504)
Signed-off-by: Yifan Qiao <yifanqiao@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Codex <noreply@openai.com>
2026-10-03 06:26:12 +00:00
f7999d2e44 [Bugfix][HiSparse] Size the KV cache from the groups HiSparse allocates (#59450)
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
2026-10-02 04:11:55 +00:00
ca65eb67d9 [Perf][HiSparse] Avoid repeated prefix scans and residency updates (#57930)
Signed-off-by: S1ro1 <matej.sirovatka@gmail.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
2026-10-01 21:48:18 +00:00
Matthew BonanniandClaude Opus 5.5 3a69636645 [Bugfix][HiSparse] Fix a chunked-prefill preemption livelock (#59494)
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-01 13:02:31 -04:00
cb6458c5ce [Core] Remove deprecated mamba_cache_mode "all" (#58997)
Signed-off-by: zjy0516 <riverclouds.zhu@qq.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Thomas Parnell <tpa@zurich.ibm.com>
2026-10-01 02:59:37 -07:00
70be6aae19 [Feature][Spec Decode] Support sampling mask replay for MRV2 MTP (#59359)
Signed-off-by: Cuiping Cheng <chengcuipingswu@163.com>
Signed-off-by: aoshen02 <aoshen@inferact.ai>
Co-authored-by: Cuiping Cheng <chengcuipingswu@163.com>
Co-authored-by: aoshen02 <aoshen@inferact.ai>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-01 13:51:39 +08:00
Nick HillandClaude Opus 5.5 c055c1e075 [Core] Include the LoRA path in prefix-cache block hashes (#59335)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-01 08:37:09 +08:00
kliuaeandNick Hill 7e583e615c [Bugfix][Core] Fix mamba prefill checkpoint block reservation and prompt-end eviction in align mode (#59175)
Signed-off-by: kliuae <kuanfu.liu@embeddedllm.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
2026-09-30 21:26:54 +00:00
Matthew BonanniClaude Opus 5.5mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
3eb6cec22a [Bugfix][HiSparse] Adopt GPU prefix copies after the hit's allocation (#59282)
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-09-30 15:43:09 -04:00
Yueh-Ting (eop) ChenandMatthew Bonanni ff1b87cca2 [Bugfix][HiSparse] Preserve host prefix publication after request completion (#59007)
Signed-off-by: Yueh-Ting Chen <yueh.ting.chen@gmail.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
2026-09-30 16:14:51 +00:00
Giulio De PasqualeandIsotr0py 2e52a558f0 [Bugfix][Core] Schedule encoder-only prompts larger than one step (#59029)
Signed-off-by: Giulio De Pasquale <git@depasquale.giugl.io>
Co-authored-by: Isotr0py <2037008807@qq.com>
2026-09-30 22:04:28 +08:00
Varun Sundar Rabindranath 5463fe4962 [KV-Offloading][TP] : Expand replicated_layout detection to multi-group MLA (#57652) 2026-09-30 09:11:03 -04:00
90e13fc757 [Bugfix][HiSparse] Never allocate GPU pages without host backing (#59036)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-30 04:13:29 +00:00
Mia Ojeda e006d761a5 [Bugfix][CPU][DiffusionGemma] Support narrower canvas w/ sync scheduling (#59107)
Signed-off-by: ojeda-e <estefania.b.ojeda@gmail.com>
2026-09-29 23:28:50 -04:00
765872e7ed [Core][BugFix] Tag prefix-cache extra keys by source (#51899)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Lucas Bourtoule <35483370+dhalf@users.noreply.github.com>
Co-authored-by: Tai An <antai12232931@outlook.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-09-29 18:24:59 -07:00
d882bddbea [Bugfix][Mamba] Keep the prompt-end prefill checkpoint under sparse retention (#59146)
Signed-off-by: Jared Wen <jaredwen@inferact.ai>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
2026-09-29 21:00:16 +00:00
aoshen02andOpenAI Codex 4861833ae2 [Bugfix][Core] Allow AuxOutput reset after abort without running requests (#59060)
Signed-off-by: aoshen02 <aoshen@inferact.ai>
Co-authored-by: OpenAI Codex <codex@openai.com>
2026-09-29 09:23:32 +00:00
Nick HillandRobert Shaw 42dbe4578c [Core] Rework scheduler skipped_waiting queue (#58947)
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
2026-09-28 20:03:16 +00:00
yzong-rh 2f4a975901 [Bugfix] Fix resumable request + async scheduling handoff race (#58259)
Signed-off-by: Yifan Zong <yzong@redhat.com>
2026-09-28 15:26:57 -04:00
927c87b347 [PCP][DCP] Support DCP target model with non-DCP Dspark (#56723)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Codex <noreply@openai.com>
2026-09-26 18:12:35 -04:00
lzOpenAI CodexLucas WilkinsonRobert Shawmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
29468dde8b [Bugfix][KV Cache][MLA] Align packed block strides for V3.2 sparse MLA (#55528)
Signed-off-by: lz <145014769+200lz@users.noreply.github.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-09-25 05:33:36 +00:00
d5051abaf1 [Bugfix][Mamba] Restore prompt-tail prefix-cache hits with MTP (#58368)
Signed-off-by: Netanel Haber <58652339+netanel-haber@users.noreply.github.com>
Co-authored-by: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Co-authored-by: Benjamin Chislett <bchislett@nvidia.com>
2026-09-24 23:46:42 +00:00
f6aa2919bc [Bugfix][Core] Keep every multimodal feature in the partial-block KV event (#58288)
Signed-off-by: haosenwang1018 <haosenwang1018@users.noreply.github.com>
Co-authored-by: haosenwang1018 <haosenwang1018@users.noreply.github.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
2026-09-24 23:25:01 +00:00
Jellow e6c07ea576 [Bugfix][KV Cache] Fix incremental multimodal block hashing (#51694)
Signed-off-by: Jellow <49915976+CZT0@users.noreply.github.com>
Signed-off-by: Jellow <dvdx@foxmail.com>
2026-09-24 15:38:03 -07:00
Wentao Ye 3c42385d7d [Refactor] Remove dead or duplicate tests (#58446)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
2026-09-24 10:29:56 -04:00
jackLeiandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com> bbc4ddeca9 [Bugfix][V1] Reject encoder-cache hits with mismatched embedding counts (#57696)
Signed-off-by: jackLei0901 <42642542+jackLei0901@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-09-24 08:40:47 +00:00
Robert Shaw 6697a7dd30 [Scheduler] Tune --long-prefill-token-threshold adaptiveness (#58459)
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
2026-09-23 22:23:18 -04:00
Andreas KaratzasandOpenAI Codex 0f2a15c927 [Tests] Select V2 for diffusion scheduler unit tests (#58272)
Signed-off-by: Andreas Karatzas <Andreas.Karatzas@amd.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
2026-09-23 02:04:22 -05:00
1b3b88ec2b [Core] structured generation mode for DiffusionGemma model (Jev-like) (#57250)
Signed-off-by: Matt Mastracci <matthew@mastracci.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Razorback16 <razorback16@protonmail.com>
Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Codex <noreply@openai.com>
Co-authored-by: Razorback16 <razorback16@protonmail.com>
2026-09-22 20:51:17 +00:00
Navjot SinghandCodex 0bce411a07 [Bugfix] Annotate MTP draft KV cache groups positionally on the hybrid grouping path (#55390)
Signed-off-by: Navjot Singh <navjot.singh@shopify.com>
Co-authored-by: Codex <noreply@openai.com>
2026-09-22 00:31:26 +00:00
Robert ShawandRobert Shaw c64b15cde5 [Scheduler] Soften Long Prefill Tokens Threshhold (#57951)
Signed-off-by: Robert Shaw <robertgshaw2@gmail.com>
Signed-off-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
2026-09-21 17:19:42 -07:00
0b7f11a1ee [AuxOutput] Add block-keyed storage for routed-expert outputs (#45635)
Signed-off-by: aoshen02 <aoshen@inferact.ai>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: aoshen02 <aoshen@inferact.ai>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Roger Wang <hey@rogerw.io>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Codex <noreply@openai.com>
2026-09-21 10:25:05 +00:00