huangtingwei
|
00fefef16b
|
[PD & HiSparse] Add DeepSeek V4 support for HiSparse direct Prefill-to-Decode DRAM (#24880)
|
2026-06-05 15:39:48 +08:00 |
|
 ![github-actions[bot]](/assets/img/avatar_default.png)
|
6cbc035dc9
|
FrozenKVMTPVerifyInput: add _draft_preprocess_idle call for when all requests in the verify batch finish in the same iteration (#26859)
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
Co-authored-by: Harmya Bhatt <harmyacs@gmail.com>
|
2026-06-04 21:47:32 -07:00 |
|
liuxianglong17
|
aed0808e18
|
6-5 nightly failed test case fix (#27335)
|
2026-06-05 11:39:23 +08:00 |
|
Yuhao Yang
|
46c58b5c70
|
bench: fix MMMU VLM eval max_tokens for CoT prompt (#27327)
|
2026-06-05 10:28:16 +08:00 |
|
   
|
2c8357f794
|
[XPU] Enable Gemma 4 E2B / E4B / 31B/ 26B-A4B on Intel XPU (#23280)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: jmunetong <jmunetong@users.noreply.github.com>
Co-authored-by: Meng, Hengyu <hengyu.meng@intel.com>
Co-authored-by: ckvermaAI <ckverma@habana.ai>
|
2026-06-05 10:05:07 +08:00 |
|
Zhangheng
|
631db6c757
|
[UnifiedTree]: Sync sidecar component hits across TP ranks and make SWA prefetch all-or-nothing (#27264)
|
2026-06-05 09:23:54 +08:00 |
|
 Cheng WanandClaude Opus 4.8
|
7dc7376697
|
fix(attn): delegate init_mha_chunk_metadata in HybridLinearAttnBackend (#27316)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-04 17:44:23 -07:00 |
|
 Cheng WanandClaude Opus 4.8
|
0aa72a9e76
|
Replace skip_attn_backend_init with a batch-carried attention plan marker (+ staleness re-plan) (#27193)
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
|
2026-06-04 17:13:18 -07:00 |
|
Lianmin Zheng
|
448d3afb76
|
fix(spec): complete CustomSpecAlgo duck-typing interface and guard against drift (#27300)
|
2026-06-04 15:52:46 -07:00 |
|
 Bowen WangandXinyuan Tong
|
07f326c184
|
Fix multimodal synthetic benchmark prompt generation to exclude special tokens (#26864)
Co-authored-by: Xinyuan Tong <115166877+JustinTong0323@users.noreply.github.com>
|
2026-06-04 22:27:43 +00:00 |
|
Liangsheng Yin
|
cd98d97037
|
Use level-1 (quiet) busy memory check in chunked-prefill and streaming tests (#27303)
|
2026-06-04 17:04:27 -04:00 |
|
Liangsheng Yin
|
687cfe9198
|
Enable runtime busy memory check for speculation topk>1 (#27228)
|
2026-06-04 16:46:20 -04:00 |
|
YC Yen-Ching Tseng
|
69623f4b11
|
[AMD] Guard aiter greedy_sample OOB token id (fixes VLM MMMU CI) (#27247)
|
2026-06-04 12:53:58 -07:00 |
|
 Chengze FanandHaiShaw
|
8e836e7dc9
|
Support optional kwargs in AITER fused_moe runner (#26746)
Co-authored-by: HaiShaw <hixiao@gmail.com>
|
2026-06-04 08:25:03 -07:00 |
|
liuxianglong17
|
8933ec8772
|
fix test cases failed on 5/30 in nightly pipeline (#26775)
|
2026-06-04 20:07:20 +08:00 |
|
liuxianglong17
|
4167f211b1
|
Solving the problem of test case failures caused by timeouts (#27027)
|
2026-06-04 20:07:04 +08:00 |
|
liuxianglong17
|
364c2bfea8
|
Test case restoration in the full test. (#26933)
|
2026-06-04 20:06:44 +08:00 |
|
shuwenn
|
1af53f67b2
|
[mem_cache][2/N] refactor: move SWATokenToKVPoolAllocator to allocator/swa.py (#26676)
|
2026-06-04 15:10:37 +08:00 |
|
YC Yen-Ching Tseng
|
6dcd78a37f
|
[AMD] Add MiniMax-M2.5 TP=4 nightly accuracy test for MI355X (#27126)
|
2026-06-03 22:17:40 -07:00 |
|
 
|
5c8a04ac4e
|
[XPU CI] Expand stage-a and consolidate stage-b tests into stage-a (#27156)
Co-authored-by: vshekhawat-hlab <vshekhawat@habana.ai>
Co-authored-by: Ma Mingfei <mingfei.ma@intel.com>
|
2026-06-04 12:00:25 +08:00 |
|
 Cheng WanandClaude Opus 4.8
|
10ab7c919f
|
[refactor] Retire DecodeInputBuffers / PrefillInputBuffers in favor of CudaGraphBufferRegistry (#27192)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-03 20:52:56 -07:00 |
|
 Zaili WangandMa Mingfei
|
3b7a258f63
|
[CPU] upgrade dependent torch ver to PT2.12 (#21456)
Co-authored-by: Ma Mingfei <mingfei.ma@intel.com>
|
2026-06-04 11:04:11 +08:00 |
|
 Zhanghengand晟海
|
736263f3dc
|
[UnifiedTree]: Support l3 storage for swa and deepseek v4 (#26881)
Co-authored-by: 晟海 <huangtingwei.htw@antgroup.com>
|
2026-06-04 10:17:34 +08:00 |
|
shuwenn
|
1a57145975
|
[codex] Fix adaptive metrics test flake (#27135)
|
2026-06-03 19:07:17 -07:00 |
|
 Yuzhen ZhouandJiajun Li
|
e03dfa8182
|
[3/N][Sync sglang-miles] TITO Support (#23751)
Co-authored-by: Jiajun Li <48857426+guapisolo@users.noreply.github.com>
|
2026-06-03 21:45:33 -04:00 |
|
 
|
f6cd1a9822
|
Add num_waiting_uncached_tokens load metric (#27174)
Co-authored-by: cctry <cctry@fb.com>
Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>
|
2026-06-03 18:29:49 -07:00 |
|
 ybyangandLianmin Zheng
|
687baf9471
|
fix(load-snapshot): avoid duplicate zmq bind in multi-tokenizer mode (#27145)
Co-authored-by: Lianmin Zheng <lianminzheng@gmail.com>
|
2026-06-03 18:24:56 -07:00 |
|
 
|
14ed9b448e
|
Add ZMQ IPv6 support, bench_serving sampling params, and reduce routed_dp_rank log noise (#27180)
Co-authored-by: Hanming Lu <69857889+hanming-lu@users.noreply.github.com>
Co-authored-by: Grigory Sizov <grisha.sizov@gmail.com>
|
2026-06-03 17:49:34 -07:00 |
|
 Clintandclintg6
|
cfb7fb4fad
|
[AMD] Fix TP2 DeepSeek-R1 nhead=64 MLA decode crash and add nightly coverage (#27188)
Co-authored-by: clintg6 <7388379+clintg6@users.noreply.github.com>
|
2026-06-03 16:56:05 -07:00 |
|
 Cheng WanandClaude Opus 4.8
|
c9ca56da8c
|
Unify full→SWA index translation in init_forward_metadata; drop pool caches (#27091)
Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com>
|
2026-06-03 16:12:27 -07:00 |
|
Cheng Wan
|
61aa3293d3
|
Revert "Fix TokenizerManager crash on top_logprobs with tensor values" (#27187)
|
2026-06-03 14:53:28 -07:00 |
|
ishandhanani
|
978fb6ed1a
|
hicache kv events: publish split write-through fragments (#27072)
|
2026-06-03 14:52:25 -07:00 |
|
Kevin Flansburg
|
7716fa00e0
|
Fix TokenizerManager crash on top_logprobs with tensor values (#26825)
|
2026-06-03 13:55:02 -07:00 |
|
YC Yen-Ching Tseng
|
d1bc06b63b
|
[AMD] Disable AITER custom all-gather in DeepSeek-R1-MXFP4 8-GPU test (#27163)
|
2026-06-03 13:38:34 -07:00 |
|
 
|
293816ab14
|
[AMD][MXFP4] Online MXFP4 quantization 1/N - dense and MOE models w. original BF16 weight (#18005)
Co-authored-by: Bowen Bao <bowenbao@amd.com>
Co-authored-by: Colin Zeng <Colin.Zeng@amd.com>
|
2026-06-03 12:55:24 -07:00 |
|
 Hanming LuandYAMY
|
e0b692600f
|
[Mamba] extra buffer lazy support (#27118)
Co-authored-by: YAMY <74099316+YAMY1234@users.noreply.github.com>
|
2026-06-03 12:42:11 -07:00 |
|
 Liangsheng YinandAlison Shao
|
ac99794e64
|
Reland spec v2 tree drafting (eagle topk>1) with page_size==1 (#26866) (#26997)
Co-authored-by: Alison Shao <54658187+alisonshao@users.noreply.github.com>
|
2026-06-03 15:40:05 -04:00 |
|
Liangsheng Yin
|
578f232e5e
|
Fix trace_modules gate disabling default trace contexts (#27173)
|
2026-06-03 14:03:42 -04:00 |
|
 Cheng WanandClaude Opus 4.8
|
45604a0f4a
|
[refactor] Unify CUDA graph runner input buffers behind CudaGraphBufferRegistry (#26742)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-03 10:54:10 -07:00 |
|
 Lijuan TangandXiaodong Ye
|
9d0e6a2df4
|
fix(mlx): set canary_manager and materialize overlap-loop inputs on Apple Silicon (#26882)
Signed-off-by: Xiaodong Ye <yeahdongcn@gmail.com>
Signed-off-by: LijuanTang94 <tang.lij@northeastern.edu>
Co-authored-by: Xiaodong Ye <yeahdongcn@gmail.com>
|
2026-06-04 00:03:46 +08:00 |
|
Ye (Charlotte) Qi
|
03c77dc33d
|
[PD] Deduplicate PD logprob normalization (#27085)
|
2026-06-03 19:08:15 +08:00 |
|
Bingxu Chen
|
d7013b6537
|
[AMD] [CI] Remove hardcoded model/cache paths from MI35x nightly tests (#27001)
|
2026-06-03 02:14:32 -07:00 |
|
 
|
e67810bea7
|
[SGLang Tracing] Add pd disaggregation mooncake backend tracing (#23755)
Co-authored-by: Mu Huai <tianbowen.tbw@antgroup.com>
Co-authored-by: Shangming Cai <csmthu@gmail.com>
|
2026-06-03 16:43:29 +08:00 |
|
 Vladislav NosivskoyandZhangheng
|
63dc20ae6c
|
[UnifiedTree] Add CP sync (#25395)
Co-authored-by: Zhangheng <hzh0425@apache.org>
|
2026-06-03 16:10:27 +08:00 |
|
Bingxu Chen
|
8e77af1afc
|
[AMD] fix(triton-mla): cap max_kv_splits at 256 on gfx942 (Kimi-K2.6 hang) (#24762)
|
2026-06-03 00:13:18 -07:00 |
|
Kevin Flansburg
|
52f2fe456a
|
fix(disagg): correct DSA/SWA state-page transfer mismatch in PD disaggregation (#27004)
|
2026-06-03 14:33:41 +08:00 |
|
 Khoa PhamandClaude Opus 4.8
|
6d53615699
|
[Gemma4] Use hard GSM8K accuracy floor for 31B MTP test (#27101)
Co-authored-by: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
|
2026-06-03 01:57:12 -04:00 |
|
Shaun Kotek
|
b8d7351a74
|
Feat/add w4a16 moe support to nemotron (#25655)
|
2026-06-02 22:42:26 -07:00 |
|
gaopengff
|
aa510bda45
|
Support specific pass of bias_grouped_topk for xpu (#26349)
|
2026-06-03 13:13:48 +08:00 |
|
 
|
f4e7a98fe5
|
[HiCache] feat: support draft offload for mooncake (#24984)
Co-authored-by: huangtingwei9988 <141888744+huangtingwei9988@users.noreply.github.com>
Co-authored-by: stmatengss <11641725+stmatengss@users.noreply.github.com>
|
2026-06-02 21:42:04 -07:00 |
|