Michael Goin and mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
e3fe212eaf
[Bugfix] Generalize KV block zeroing to AttentionSpec ( #51749 )
...
Signed-off-by: mgoin <mgoin64@gmail.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-08-11 12:13:09 -04:00
Nick Hill
513f83e7ec
[Bugfix] Take the sliding window from the layer, not the KV cache group ( #51756 )
...
Signed-off-by: Nick Hill <nickhill123@gmail.com >
2026-08-11 09:10:46 -07:00
Harry Mellor
c65aa2ee6c
Bump Transformers version to 5.15.0 ( #51668 )
...
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com >
2026-08-11 09:00:51 -07:00
Guan-Ming Chiu and mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
457a5f34db
[Bugfix][MRV2] Support encoder timing stats in model runner V2 ( #50020 )
...
Signed-off-by: Guan-Ming (Wesley) Chiu <105915352+guan404ming@users.noreply.github.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-08-12 00:00:49 +08:00
5426311d91
[Kernel][ROCm][Perf] FlyDSL decode-attention kernel for 4-bit TurboQuant KV cache ( #47896 )
...
Signed-off-by: root <root@smci355-ccs-aus-m02-09.cs-aus.dcgpu >
Co-authored-by: root <root@smci355-ccs-aus-m02-09.cs-aus.dcgpu >
Co-authored-by: Andreas Karatzas <akaratza@amd.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-08-11 08:45:39 -07:00
liuzhenwei and Kunshang Ji
d648236199
[XPU][CI] Fix ExampleConnector KV cache device selection ( #51806 )
...
Signed-off-by: zhenwei-intel <zhenwei.liu@intel.com >
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com >
2026-08-11 23:40:29 +08:00
pmanczak and Kunshang Ji
5b5eae2394
[Misc] Enable test_silu_mul_fp8_quant_deep_gemm on XPU ( #49444 )
...
Signed-off-by: pmanczak <pawel.manczak@intel.com >
Co-authored-by: Kunshang Ji <kunshang.ji@intel.com >
2026-08-11 23:19:14 +08:00
Wentao Ye
4988df2eb8
[Config] Update default _max_num_batched_tokens from 8192 to 16384 ( #51726 )
...
Signed-off-by: yewentao256 <zhyanwentao@126.com >
2026-08-11 11:09:34 -04:00
Woosuk Kwon and OpenAI Codex
87668ab69b
[Bugfix] Guard DeepSeek V4 MRV1 piecewise CUDA graphs ( #51768 )
...
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai >
Co-authored-by: OpenAI Codex <codex@openai.com >
2026-08-11 07:34:44 -07:00
b64a2708b0
Bound generation inputs before expensive work ( #51447 )
...
Signed-off-by: Clinton Thomas <1033162+KernelClint@users.noreply.github.com >
Co-authored-by: Lucas Bourtoule <35483370+dhalf@users.noreply.github.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-08-11 22:03:50 +08:00
6c95a641e9
[2/N][Feat][Perf] Add new warmup infrastructure for JITs. Add predicate filtering for JIT warmup, and migrate Inkling FA4 ( #49315 )
...
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com >
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com >
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-08-11 07:46:07 -04:00
AlexHuang
0fec3d652b
[Bugfix][KV Offload] Centralize shared mmap cleanup in CPU worker ( #51622 )
...
Signed-off-by: Alex <jihui.huang@daocloud.io >
2026-08-11 13:22:15 +03:00
Dao007forever and OpenAI Codex
ce07118669
[Bugfix][Core] Preserve Mamba running CoW after external hits ( #51766 )
...
Signed-off-by: Dao Le <Dao007forever@gmail.com >
Co-authored-by: OpenAI Codex <codex@openai.com >
2026-08-11 18:15:24 +08:00
Tianyu Guo and OpenAI Codex
3fb7bb46d6
[Perf] Avoid repeated multimodal prompt update scans ( #51774 )
...
Signed-off-by: Tianyu Guo <guoty@inferact.ai >
Co-authored-by: OpenAI Codex <codex@openai.com >
2026-08-11 09:23:24 +00:00
Yizhou
a311916a29
[MRV2][Spec] Fuse AR speculator multi-step decodes back into one CUDA graph ( #46849 )
...
Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com >
2026-08-11 15:06:22 +08:00
Fangzhou Ai
1044d19998
[ROCm][DSV4] Preserve native MXFP4 TP8 shard allocation ( #51473 )
...
Signed-off-by: Fangzhou Ai <fangzhou.ai@amd.com >
2026-08-10 23:57:27 -07:00
Priyansh Jain and Andreas Karatzas
2acb055ed7
[CPU][Zen] Route BF16 MoE inference through zentorch on AMD ( #44201 )
...
Co-authored-by: Andreas Karatzas <akaratza@amd.com >
2026-08-10 23:20:05 -07:00
Wei Zhao
608c12473f
[Attention] Fix MLA prefill workspace allocation size ( #51733 )
...
Signed-off-by: wzhao18 <wzhao18.sz@gmail.com >
2026-08-11 05:34:30 +00:00
c76a425278
[Kernel] Optimize long-context MLA cache gathers ( #51739 )
...
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com >
Co-authored-by: OpenAI Codex <noreply@openai.com >
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com >
2026-08-10 22:23:43 -07:00
cherry77-cloud and OpenAI Codex
65b7662d3f
[Bugfix][Frontend] Report Cohere stop sequences correctly ( #51556 )
...
Signed-off-by: cherry77-cloud <1615405@qq.com >
Co-authored-by: OpenAI Codex <codex@openai.com >
2026-08-11 04:49:49 +00:00
Khushali Desai and wang.yuqi
dc5101fb1b
replace batch_norm to numerically identical without cudnn ( #51734 )
...
Signed-off-by: khushali9 <khushali.desai9@gmail.com >
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io >
Co-authored-by: wang.yuqi <yuqi.wang@daocloud.io >
2026-08-11 12:43:20 +08:00
Clinton Thomas and Lucas Bourtoule
90fd4a333f
Preserve revision pins in secondary artifact loaders ( #51446 )
...
Signed-off-by: Clinton Thomas <1033162+KernelClint@users.noreply.github.com >
Co-authored-by: Lucas Bourtoule <35483370+dhalf@users.noreply.github.com >
2026-08-11 04:33:10 +00:00
Frank
adc1200597
[Misc] Use VLLMValidationError in scoring input validation ( #51753 )
...
Signed-off-by: frank-suwen <suwenw2@outlook.com >
2026-08-11 04:12:22 +00:00
Wentao Ye
0914ed2e81
[Perf] Adaptive budget for spec scheduled input tokens, ~60% better Kimi K3 DSpark TTFT ( #51725 )
...
Signed-off-by: yewentao256 <zhyanwentao@126.com >
2026-08-10 20:52:21 -07:00
Itay Etelis and Itay Etelis
0f2ea973be
[KV Connector][Offloading] Keep per-layer KV registration when canonical_layout is requested ( #51688 )
...
Signed-off-by: Itay Etelis <itay.etelis@ibm.com >
Co-authored-by: Itay Etelis <itay.etelis@ibm.com >
2026-08-11 06:03:59 +03:00
aoshen02 and Roger Wang
99e62b802c
[Bugfix][Model Loader] Defer post-load attention weight processing ( #49519 )
...
Signed-off-by: aoshen02 <aoshen@inferact.ai >
Co-authored-by: Roger Wang <hey@rogerw.io >
2026-08-10 19:16:57 -07:00
33584901ba
[Frontend] Add content_parts to /inference/v1/generate for raw multim… ( #51478 )
...
Signed-off-by: Allen Shen <aoshen@inferact.ai >
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-08-11 10:00:12 +08:00
Andreas Karatzas and OpenAI Codex
1a1727330a
[CI] Solidify speculative decoding E2E coverage ( #50713 )
...
Signed-off-by: Andreas Karatzas <Andreas.Karatzas@amd.com >
Signed-off-by: Andreas Karatzas <akaratza@amd.com >
Co-authored-by: OpenAI Codex <codex@openai.com >
2026-08-10 19:56:39 -06:00
Tarun Kumar
48bada6ea4
Fix chat completion 500 on non-object JSON bodies ( #51654 )
...
Signed-off-by: Tarun Kumar <takumar@redhat.com >
2026-08-11 01:31:19 +00:00
Almog Tavor
c3cac8c63d
[Bugfix][MiMo] Apply vision attention sinks in the window attention path ( #49815 )
...
Signed-off-by: almogtavor <almogtavor@gmail.com >
2026-08-10 22:36:42 +00:00
Tristan Rice
98a4144a41
[Bugfix] Preserve non-logitproc entry points in tests ( #51097 )
...
Signed-off-by: Tristan Rice <rice@fn.lc >
2026-08-10 18:14:53 -04:00
63ac04a61e
[Kimi-K3] DCP support ( #50484 )
...
Signed-off-by: girasoley <girasoley@inferact.ai >
Signed-off-by: Summer Yang <girasoleyang@gmail.com >
Signed-off-by: 云挚 <ningyunxiao.nyx@antgroup.com >
Co-authored-by: Claude Fable 5 <noreply@anthropic.com >
Co-authored-by: 云挚 <ningyunxiao.nyx@antgroup.com >
Co-authored-by: foraxe <73625538+foraxe@users.noreply.github.com >
Co-authored-by: Woosuk Kwon <woosuk@inferact.ai >
2026-08-10 12:54:26 -07:00
Flora Feng
3f142bd85e
[Bugfix] Align deepseek v4 parser thinking default with tokenizer ( #51296 )
...
Signed-off-by: sfeng33 <4florafeng@gmail.com >
2026-08-10 19:30:30 +00:00
Flora Feng
05f0a80016
[Bugfix][Structured Output] Mask request stop tokens in xgrammar until grammar terminates ( #49227 )
...
Signed-off-by: sfeng33 <4florafeng@gmail.com >
2026-08-10 14:29:20 -04:00
Lyu, Xudong
0e2d78028c
[Bugfix][Kimi-K3] Give the AMD packed KDA decode kernel the state-index stride ( #51682 )
...
Signed-off-by: Lyu, Xudong <xudong.lyu@amd.com >
2026-08-11 00:05:20 +08:00
Juan Pérez de Algaba and Cursor
640a09086d
Fix DoS via sample-rate forgery bypassing audio decode duration guard ( #49948 )
...
Signed-off-by: jperezde <jperezde@redhat.com >
Co-authored-by: Cursor <cursoragent@cursor.com >
2026-08-10 15:49:17 +00:00
fanxingran
37fbf52084
[ROCm][MLA] [K3] Fix fp8 KV cache decode on the AITER MLA backend ( #51011 )
...
Signed-off-by: fanxingran <xingran.fan@amd.com >
2026-08-10 11:42:10 -04:00
Francesco Fusco and Claude
fac808b36f
[Perf][Hybrid] 3D-grid tiling of the state-copy Triton kernels ( #49436 )
...
Signed-off-by: Francesco Fusco <ffu@zurich.ibm.com >
Co-authored-by: Claude <noreply@anthropic.com >
2026-08-10 17:02:49 +02:00
liuzhenwei
bd6536071c
[XPU][Test] Pin block size in test_multi_connector ( #51213 )
...
Signed-off-by: zhenwei-intel <zhenwei.liu@intel.com >
2026-08-10 21:57:45 +08:00
pmanczak
ec21f61d82
[Misc] Enable test_fused_moe_wn16 on XPU ( #51672 )
...
Signed-off-by: pmanczak <pawel.manczak@intel.com >
2026-08-10 21:57:02 +08:00
Taneem Ibrahim and Wentao Ye
ea3115e30b
[CI] Stabilize DP supervisor lifecycle tests ( #51557 )
...
Signed-off-by: Taneem Ibrahim <taneem.ibrahim@gmail.com >
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com >
2026-08-10 09:13:52 -04:00
Harry Mellor
cf8f3a3bb2
[2/N] Harden Transformers modelling backend multi-modal path ( #51657 )
...
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com >
2026-08-10 12:57:12 +00:00
Raj Vijay Firke and Harry Mellor
3dafaef027
[Bugfix][Core] Emit --no-{key} for false BooleanOptionalAction flags in YAML config ( #51573 )
...
Signed-off-by: Raj Firke <79653531+rajfirke@users.noreply.github.com >
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com >
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com >
2026-08-10 11:05:15 +00:00
efschu
900d09f91a
[Bugfix][Model] Fix Qwen3.5 MTP for text-only checkpoints ( #50734 )
...
Signed-off-by: efschu <51944948+efschu@users.noreply.github.com >
2026-08-10 11:04:53 +00:00
243c63baf5
[V1][Scheduler] Apply Mamba alignment before encoder caps ( #51603 )
...
Signed-off-by: Jiangyun Zhu <riverclouds.zhu@qq.com >
Co-authored-by: Akshat Anand <40275336+cipheraxat@users.noreply.github.com >
Co-authored-by: Codex <codex@openai.com >
2026-08-10 16:37:17 +08:00
Harry Mellor
70b84f0bcb
Fix Gemma 4 for upcoming Transformers version ( #49797 )
...
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com >
2026-08-10 07:48:26 +00:00
Itay Etelis and Itay Etelis
81840a172f
[KV Connector] Canonical CPU layout for parallelism-agnostic KV offload ( #48414 )
...
Signed-off-by: Itay Etelis <itay.etelis@ibm.com >
Co-authored-by: Itay Etelis <itay.etelis@ibm.com >
2026-08-10 10:44:54 +03:00
Andreas Karatzas and Micah Williamson
31cd109f18
[ROCm][CI] Extend ROCm AITER MHA (FA) coverage ( #40958 )
...
Signed-off-by: Andreas Karatzas <akaratza@amd.com >
Signed-off-by: Micah Williamson <micah.williamson@amd.com >
Co-authored-by: Micah Williamson <micah.williamson@amd.com >
2026-08-10 15:35:13 +08:00
Zetian Li - ikun and Claude
7303c66f68
[Bugfix] Fix lfm2 tool parser dropping calls with brackets or newline… ( #48171 )
...
Signed-off-by: Zetian Li <804561096@qq.com >
Co-authored-by: Claude (Anthropic) <noreply@anthropic.com >
2026-08-10 15:10:06 +08:00
61c1dd0966
[EC Connector] Call to EC Connector update_connector_output from scheduler ( #49579 )
...
Signed-off-by: omerpaz95 <omerpaz95@gmail.com >
Signed-off-by: Nicolò Lucchesi <nicolo.lucchesi@gmail.com >
Co-authored-by: Nicolò Lucchesi <nicolo.lucchesi@gmail.com >
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
2026-08-10 14:25:57 +08:00
