 Rahul ChalamalaandJanelle Cai
|
03a8d0b1ed
|
[Model][Spec Decode] Tap the pre-norm AttnRes mixture as the Kimi K3 DFlash aux state (#50487)
Signed-off-by: Rahul Chalamala <22563365+rchalamala@users.noreply.github.com>
Co-authored-by: Janelle Cai <janelle.cai@modal.com>
|
2026-08-14 09:03:42 -07:00 |
|
|
Lucas WilkinsonandClaude
|
57bd0ed441
|
[5/N][KV-Cache Layout Refactor] Backend-published KV packing via customize_spec (#51704)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Claude <noreply@anthropic.com>
|
2026-08-14 09:47:55 -04:00 |
|
|
fxmarty-amd
|
48825acc23
|
[Quantization] Remove dead QuantizationConfig.is_mxfp4_quant (#51793)
Signed-off-by: Felix Marty <Felix.Marty@amd.com>
|
2026-08-13 13:58:31 -04:00 |
|
|
Nick Iusiumbeli
|
83d4c6196a
|
[Bugfix] Declare SupportsEagle3 on KimiLinearForCausalLM (#52171)
Signed-off-by: Nick Iusiumbeli <nickuspro@gmail.com>
|
2026-08-13 16:22:50 +00:00 |
|
|
Thien Tran
|
6014f9e672
|
[Kimi-K3] Add GEMM-RS for sequence parallelism (#52079)
Signed-off-by: Thien Tran <gau.nernst@yahoo.com.sg>
|
2026-08-13 09:02:28 -07:00 |
|
|
范裕达
|
170592a931
|
[Bugfix] Disable sequence parallelism for Dots3 NOTE (#52172)
Signed-off-by: KurodaKanbei <mistergalahad@gmail.com>
|
2026-08-13 22:13:00 +08:00 |
|
|
Harry Mellor
|
903da602f3
|
[Docs] Fix WhisperEncoderLayer.forward docstring in dots3_note (#52134)
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
|
2026-08-13 02:32:14 -07:00 |
|
|
kliuae
|
f96261637c
|
[ROCm][Perf] Kimi-K3 Remove prefill pipeline stall in chunk KDA (#51862)
Signed-off-by: kliuae <kuanfu.liu@embeddedllm.com>
|
2026-08-13 09:11:45 +00:00 |
|
|
|
903d2efe7e
|
[Attention][MLA] Fuse Kimi-K3 chunked-context K/V packing (#51772)
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-08-12 22:42:57 -07:00 |
|
|
stefankoncarevicandAndreas Karatzas
|
b369f10d5c
|
[Bugfix][ROCm][CI] Restore the DeepSeek-V4 input GEMM override point (#51821)
Signed-off-by: Stefan Koncarevic <Stefan.Koncarevic@amd.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>
|
2026-08-12 20:33:52 -05:00 |
|
|
|
7f7a32cfec
|
[Spec Decode] DSpark confidence-scheduled verification (#47808)
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Signed-off-by: Benjamin Chislett <chislett.ben@gmail.com>
Signed-off-by: Lucas Wilkinson <wilkinson.lucas@gmail.com>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: Benjamin Chislett <chislett.ben@gmail.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: OpenAI Codex <noreply@openai.com>
|
2026-08-12 12:27:48 -07:00 |
|
|
Wentao Yeandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
fe889ac925
|
[K3 Perf] Flash kda out kernel for prefill, 1.1~1.4x kernel performance improvement (#51311)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-08-12 14:51:48 -04:00 |
|
|
Hongxia YangandClaude Fable 5
|
b745d08de1
|
[ROCm][K3] Dequantize the fp8 decode query for MLA backends without quant-query support - TRITON_MLA (#51860)
Signed-off-by: Hongxia Yang <hongxia.yang@amd.com>
Co-authored-by: Claude Fable 5 <noreply@anthropic.com>
|
2026-08-12 12:40:41 -04:00 |
|
|
|
9035151d6c
|
[Model] Add native Dots3 NOTE multimodal support (#51255)
Signed-off-by: KurodaKanbei <mistergalahad@gmail.com>
Signed-off-by: Jee Jee Li <jeejeelee@inferact.ai>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Jee Jee Li <jeejeelee@inferact.ai>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
|
2026-08-12 23:13:20 +08:00 |
|
|
aoshen02
|
4eef91c03c
|
[Model] Support R3 capture with DeepGEMM MegaMoE (#51831)
Signed-off-by: aoshen02 <aoshen@inferact.ai>
|
2026-08-12 01:07:07 -07:00 |
|
|
|
7f9173dfa2
|
[ROCm][Perf] Kimi-K3 Fused kernel for KDA decode (#50654)
Signed-off-by: kliuae <kuanfu.liu@embeddedllm.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: TJian <tunjian.tan@embeddedllm.com>
|
2026-08-11 20:39:34 -07:00 |
|
|
Wentao Ye
|
fd04edef3e
|
[Refactor] Delete dead code in models (#51838)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
|
2026-08-11 15:44:26 -04:00 |
|
|
Michael Goin
|
ded6c452c3
|
[Bugfix] Support HF-config compat for Inkling (#51850)
Signed-off-by: mgoin <mgoin64@gmail.com>
|
2026-08-11 15:32:28 -04:00 |
|
|
Tuukka Sarvi
|
12bea3eedc
|
[Bugfix][ROCm] Fix DeepSeek V4 DSpark probabilistic startup (#51145)
Signed-off-by: Tuukka Sarvi <tuukka.sarvi@amd.com>
|
2026-08-11 22:15:31 +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 |
|
|
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 |
|
|
|
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 |
|
|
|
d40c3e3c00
|
Fix DSpark warmup without sparse index buffer (#50693)
Signed-off-by: xijiade.aihemaiti <3146335281@qq.com>
Signed-off-by: xijiaat <52128022+xijiaat@users.noreply.github.com>
Co-authored-by: Wentao Ye <44945378+yewentao256@users.noreply.github.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
|
2026-08-10 15:12:19 -04:00 |
|
|
Woosuk KwonandOpenAI Codex
|
79c865b838
|
[Perf] Narrow DeepSeek V4 eager CUDA graph region (#51430)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
Co-authored-by: OpenAI Codex <codex@openai.com>
|
2026-08-10 10:52:10 -07:00 |
|
|
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 |
|
|
Jeff (Junze) Ma
|
04d13b5d65
|
[K3] Allow tpu to import kimi_k3.common (#51529)
Signed-off-by: Jeff Ma <jeffjma@umich.edu>
|
2026-08-09 17:06:34 +08:00 |
|
|
Woosuk Kwon
|
e644c8cd8c
|
[Perf] Optimize DeepSeek V3.2 sequence parallelism (#51434)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
|
2026-08-07 16:51:01 -07:00 |
|
|
Woosuk Kwon
|
46b5864054
|
[Perf] Narrow DeepSeek V3.2 eager CUDA graph region (#51425)
Signed-off-by: Woosuk Kwon <woosuk@inferact.ai>
|
2026-08-07 14:37:43 -07:00 |
|
|
Wentao Ye
|
4a11ed5bc4
|
[DSv32/GLM Perf] Skip short prefill topk for dense mha layer, 97.9% kernel level latency reduction (#51298)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
|
2026-08-07 20:30:42 +00:00 |
|
|
Wentao Yeandmergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
56a4b63d44
|
[K3 Perf] Optimize k3 dspark fused kv, 4.5~4.6x kernel performance improvement (#50585)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-08-07 12:16:11 -07:00 |
|
|
Wentao Ye
|
c84789c40b
|
[Refactor] Remove kernel dead code (#51051)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
|
2026-08-07 01:20:19 -07:00 |
|
|
|
da788334bc
|
Support DeepSeek-V4 AMD Quark NVFP4 with emulation kernel (#47972)
Signed-off-by: Jimmy Adams <Adam_ji@outlook.com>
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
Signed-off-by: huji <huidong.ji@amd.com>
Signed-off-by: jimmy-adams <41593649+jimmy-adams@users.noreply.github.com>
Co-authored-by: Andreas Karatzas <akaratza@amd.com>
Co-authored-by: huji <huidong.ji@amd.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
|
2026-08-07 01:19:31 -07:00 |
|
|
kliuae
|
43d691ec6b
|
[ROCm][Perf] Kimi-K3 Shard Latent MoE up-projection for ROCm path (#51253)
Signed-off-by: kliuae <kuanfu.liu@embeddedllm.com>
|
2026-08-07 06:11:18 +00:00 |
|
|
|
b38e111d3e
|
[Attention][MLA] Per-request scheduling for MLA chunked context (#50613)
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>
Co-authored-by: OpenAI Codex <noreply@openai.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-08-06 16:45:44 +00:00 |
|
|
yjzandJianDan0212
|
5fba75aefe
|
[Bugfix][Model] Add missing fused_qkv_a_proj to Kimi-Linear packed_modules_mapping (#51249)
Signed-off-by: JianDan0212 <zhangyj0212@gmail.com>
Co-authored-by: JianDan0212 <zhangyj0212@gmail.com>
|
2026-08-06 11:39:17 +00:00 |
|
|
Wentao Ye
|
ad5280255b
|
[GLM Perf] DSv32/glm use skip topk for MTP case, 2.0x kernel performance improvement (#50904)
|
2026-08-06 01:48:36 -07:00 |
|
|
|
2dfb8ba590
|
[Perf][CUDA] Programmatic dependent launch for the DSA decode kernels (#50230)
Signed-off-by: Peiyuan Zhou <peiyuanzhou1994@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-08-06 00:53:17 -07:00 |
|
|
Jee Jee Li
|
f85c1d2f84
|
K3: remove the add operation for megamoe path (#51146)
Signed-off-by: Jee Jee Li <jeejeelee@inferact.ai>
|
2026-08-06 01:48:25 +00:00 |
|
|
Wentao Ye
|
877975897d
|
[K3 Perf] Combine multiple all gather together for SP, 1.5~3x kernel level performance improvement (#51070)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
|
2026-08-05 17:42:41 -04:00 |
|
|
Ziming Huang
|
beca88e59e
|
[BugFix][K3] Skip moe_intermediate padding when EP is enabled (#51131)
|
2026-08-05 09:22:55 +00:00 |
|
|
kliuae
|
f5cd862dbd
|
[ROCm][Bugfix] Kimi-K3 Fix KDA NaN on mixed batches and racy autotune config (#50649)
Signed-off-by: kliuae <kuanfu.liu@embeddedllm.com>
|
2026-08-05 08:07:21 +00:00 |
|
|
Dr Tasos VaroudisandLucas Wilkinson
|
c416f15710
|
[Model] Fix Kimi-K3 MLA with disabled context parallelism (#50404)
Signed-off-by: Tasos Varoudis <varoudis@archtech.gr>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
|
2026-08-05 02:33:14 +00:00 |
|
|
Andreas Karatzas
|
33c50587d2
|
[ROCm] Restore Inkling MTP backend parity (#50806)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
|
2026-08-05 02:02:57 +00:00 |
|
|
xiaozhoupyandClaude
|
3756bf1e2b
|
[Kernel][SM100] Add a CuTeDSL fused query kernel (#49792)
Signed-off-by: Peiyuan Zhou <peiyuanzhou1994@gmail.com>
Co-authored-by: Claude <noreply@anthropic.com>
|
2026-08-04 15:23:06 -07:00 |
|
|
Canlin Guo
|
edbc4969a7
|
[Kernel][Inkling] Fuse shared-expert partial addition into the Lamport collective (#50697)
Signed-off-by: Canlin Guo <canlinguosdu@gmail.com>
|
2026-08-04 18:41:35 +00:00 |
|
|
Wentao Ye
|
d31de3c421
|
[Kimi K3 Perf] option to shard the shared expert for non mega case, 16.98 GiB memory/GPU saved (#50912)
Signed-off-by: yewentao256 <zhyanwentao@126.com>
|
2026-08-04 14:00:44 -04:00 |
|
|
|
7ac2ec7582
|
[Kimi-K3][AMD] Fuse AttnRes state updates and norms (#50593)
Signed-off-by: Liuyinfeng01 <199041580+LiuYinfeng01@users.noreply.github.com>
Co-authored-by: Liuyinfeng01 <199041580+LiuYinfeng01@users.noreply.github.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
|
2026-08-04 12:32:13 -04:00 |
|
|
namgyu-younandClaude Opus 5
|
41ba11b841
|
[Bugfix][Kimi-K3] Enforce packed rows and op availability in AttnRes dispatch (#50567)
Signed-off-by: namgyu-youn <namgyu.dev@gmail.com>
Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
|
2026-08-03 17:13:33 -07:00 |
|
|
|
c810937573
|
[Kernel] Extend CuTe DSL skinny GEMM to GLM-5.2 (#49791)
Signed-off-by: Peiyuan Zhou <peiyuanzhou1994@gmail.com>
Co-authored-by: OpenAI Codex <codex@openai.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
|
2026-08-03 16:49:41 -07:00 |
|
|
Raphaël Rialland
|
4a3447d200
|
[Bugfix][Model Runner V2] Restore multimodal draft capability detection (#50417)
Signed-off-by: Raphael Rialland <raphael.rialland@mistral.ai>
|
2026-08-03 14:18:14 -07:00 |
|