 Xiaoyu Zhang
|
0e4d1b49d3
|
[Codex] Remove stale DeepSeek V4 JIT kernels (#25764)
|
2026-05-19 20:04:32 +08:00 |
|
|
Arseniy MironovandNapkin-AI
|
45a85efc3a
|
[Diffusion][NPU]Add attention backends for diffusion models for Ascend NPU (#23482)
Co-authored-by: Napkin-AI <arseniy.mironov.dev@gmail.com>
|
2026-05-19 12:46:55 +03:00 |
|
|
Thomas
|
58b5fe3e29
|
[Diffusion] [NPU] Fix HunyuanVideo crash on NPU (#25592)
|
2026-05-19 12:40:43 +03:00 |
|
|
jianzhao-xu
|
5073c82a37
|
transformers v5 adapt HFRunner (#23922)
|
2026-05-19 17:07:38 +08:00 |
|
|
shiyu7
|
7e0818038a
|
fix: fix deepseek v4 CP error (#25396)
|
2026-05-19 02:04:21 -07:00 |
|
|
|
67fd005b97
|
[HiSparse & PD] Support hisparse memory pool host page > 1 (#23606)
Co-authored-by: hzh0425 <hzh0425@apache.org>
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
|
2026-05-19 01:29:35 -07:00 |
|
|
Ziang Li
|
78cb38ed5e
|
[FlashInfer v0.6.11] [RL] Support FlashInfer per-token NVFP4 MoE (#22918)
|
2026-05-19 01:04:48 -07:00 |
|
|
Kevin Li
|
fbfddfd5c7
|
fix (jit kernel): elementwise activation C++ error (#25695)
|
2026-05-19 15:23:52 +08:00 |
|
|
Yuhao Yang
|
79ea30d1f1
|
[Bug] Fix V4-Pro NaN on Blackwell by converting fp8_einsum input scale to ue8m0 (#25733)
|
2026-05-18 23:48:34 -07:00 |
|
|
Junlin Wuandronnie_zheng
|
4c9f31b85e
|
✨ [diffusion][npu][quant] Add MXFP4 quantization support for Wan2.2 Diffusion on Ascend NPU (#22338)
Co-authored-by: ronnie_zheng <zl19940307@163.com>
|
2026-05-19 07:46:52 +03:00 |
|
|
Hanming Lu
|
862d39e06c
|
[Mamba] Fix extra_buffer overlap schedule races (#24954)
|
2026-05-19 12:13:20 +08:00 |
|
|
Zhonghua Deng
|
f0763859ed
|
perf(mimo-v2-epd): enable GPU image preprocess and parallel video decode (#25588)
|
2026-05-19 11:47:21 +08:00 |
|
|
Yuhao Yang
|
d8e66e54e5
|
fix: use triton_attn as default vision attention on B300 (SM103) (#25570)
|
2026-05-19 11:00:07 +08:00 |
|
|
Xiaoyu Zhang
|
31e324391b
|
[Codex] Opt Mistral Large performace (#24611)
|
2026-05-19 10:59:51 +08:00 |
|
|
Mick
|
a7b3ced334
|
[diffusion] fix: fix LTX2 resident defaults and stage profiling (#25596)
|
2026-05-19 10:41:28 +08:00 |
|
|
ishandhananiandShangming Cai
|
87c3c96bc8
|
[Bug][PD][NIXL] always send aux on is_last; only expects_state when truthy (#25699)
Co-authored-by: Shangming Cai <csmthu@gmail.com>
|
2026-05-19 10:04:04 +08:00 |
|
|
huangtingweiandhzh0425
|
c2a212bfe2
|
[UnifiedTree] Support DeepSeek V4 host pool with multiple layouts. (#25282)
Co-authored-by: hzh0425 <hzh0425@apache.org>
|
2026-05-19 09:36:00 +08:00 |
|
|
Lianmin Zheng
|
b45b52ee8f
|
Add spec_verify_calls_total metric for speculative decoding (#25689)
|
2026-05-18 18:35:11 -07:00 |
|
|
fzyzcjy
|
e4d81e48c9
|
Pull the max-prefix-len computation into its own helper and rename the matched-token argument (#25728)
|
2026-05-19 09:27:06 +08:00 |
|
|
Xiaoyu ZhangandCodex
|
2424303dfb
|
[codex] Optimize hidden-size 512 RMSNorm dispatch (#24710)
Co-authored-by: Codex <codex@example.com>
|
2026-05-19 09:26:10 +08:00 |
|
|
fzyzcjy
|
170fe57cf0
|
Encapsulate the pending-flush bookkeeping in a small wrapper (#25727)
|
2026-05-19 09:23:10 +08:00 |
|
|
fzyzcjy
|
5067da3eb6
|
Confine req-pool-idx assignment to the pool allocator (#25726)
|
2026-05-19 09:22:51 +08:00 |
|
|
fzyzcjy
|
6bcbf4de35
|
Fix the misnamed request finish-check method to reflect its mutating semantics (#25725)
|
2026-05-19 09:22:28 +08:00 |
|
|
fzyzcjy
|
2d2dff28da
|
Return a mamba tracking entry from the cache lookup instead of mutating caller lists (#25724)
|
2026-05-19 09:21:55 +08:00 |
|
|
fzyzcjy
|
3fd6a58e6c
|
Inline the single-use split-prefill setup at its caller (#25722)
|
2026-05-19 09:21:29 +08:00 |
|
|
fzyzcjy
|
1f3e5aa1e0
|
Publish elastic-EP active ranks from a dedicated step (#25721)
|
2026-05-19 09:21:04 +08:00 |
|
|
fzyzcjy
|
fa37b68653
|
Rename the request mid-chunk flag to describe what it actually tracks (#25720)
|
2026-05-19 09:20:43 +08:00 |
|
|
fzyzcjy
|
32f1259c91
|
Confine max-prefix-len to where it is used and drop the leftover variable (#25719)
|
2026-05-19 09:20:21 +08:00 |
|
|
fzyzcjy
|
1cba3ab467
|
Stop returning the unused prefix-computed flag from priority calc (#25718)
|
2026-05-19 09:20:00 +08:00 |
|
|
fzyzcjy
|
2d868656d0
|
Move the retract-decode ratio estimation onto the new-token-ratio tracker (#25717)
|
2026-05-19 09:19:42 +08:00 |
|
|
fzyzcjy
|
1a882c5c63
|
Pack scattered new-token-ratio state into a dedicated tracker (#25716)
|
2026-05-19 09:19:20 +08:00 |
|
|
fzyzcjy
|
954b5c5846
|
Pack scattered scheduler IPC channel state into a dedicated container (#25714)
|
2026-05-19 09:19:02 +08:00 |
|
|
fzyzcjy
|
07b4f262b7
|
Set up the idle sleeper outside of the IPC channel initialization (#25713)
|
2026-05-19 09:18:36 +08:00 |
|
|
fzyzcjy
|
2d40f45193
|
Pack scattered request logprob state into a dedicated container (#25712)
|
2026-05-19 09:18:19 +08:00 |
|
|
fzyzcjy
|
fb7e49d4eb
|
Expose can-run-cuda-graph as a regular property on the embedding result (#25711)
|
2026-05-19 09:18:00 +08:00 |
|
|
fzyzcjy
|
a740f8de33
|
Remove the dead hasattr fallback around the test-only crash counter (#25710)
|
2026-05-19 09:17:40 +08:00 |
|
|
fzyzcjy
|
0e198f0f4f
|
Refactor batch_result_processor into per-step prefill/decode helpers (#25709)
|
2026-05-19 09:17:17 +08:00 |
|
|
fzyzcjy
|
7e7cb969e9
|
Route streaming output through the accumulator's payload method instead of an inline send (#25708)
|
2026-05-19 09:16:42 +08:00 |
|
|
fzyzcjy
|
da50e3d943
|
Log per-request time stats in a dedicated tail step (#25707)
|
2026-05-19 09:16:22 +08:00 |
|
|
fzyzcjy
|
b911fd1673
|
Route streaming-accept decisions through the accumulator instead of an inline gate (#25706)
|
2026-05-19 09:15:45 +08:00 |
|
|
fzyzcjy
|
d8f190dfba
|
Pack scattered output-streamer state into a dedicated accumulator (#25705)
|
2026-05-19 09:14:59 +08:00 |
|
|
fzyzcjy
|
e8e55bb19b
|
Split the request-reception loop into smaller phases (#25703)
|
2026-05-19 09:12:47 +08:00 |
|
|
Kurkur
|
d90bc65e30
|
[NPU] Fix TypeError in get_state_buf_infos when index_head_dim is None on MLA (#25383)
|
2026-05-19 09:09:11 +08:00 |
|
|
Ratish P
|
dbac464726
|
[Spec]: Make Triton standalone spec test deterministic (#25303)
|
2026-05-18 17:36:10 -07:00 |
|
|
Baizhou Zhangandhnyls2002
|
b79e4b1e68
|
[Fix] Try to fix error caused by latest cutedsl packages (#25690)
Co-authored-by: hnyls2002 <lsyincs@gmail.com>
|
2026-05-18 16:51:32 -07:00 |
|
|
|
314dedf7c6
|
Use SGLANG_CACHE_DIR env for gpu_p2p_access_cache path (#25686)
Co-authored-by: Ian O'Connell <ianoc@meta.com>
Co-authored-by: ianoc <ianoc@fb.com>
|
2026-05-18 16:00:54 -07:00 |
|
|
Lianmin ZhengandHanming Lu
|
745abd6cc0
|
Add no_combine support to cutlass_moe_fp4 (#25688)
Co-authored-by: Hanming Lu <69857889+hanming-lu@users.noreply.github.com>
|
2026-05-18 15:59:55 -07:00 |
|
|
|
878e6b8886
|
[SP] Fix runtime_max_tokens_per_rank for sequence parallelism (#25685)
Co-authored-by: Ming Yang <minos.future@gmail.com>
Co-authored-by: Yinghai Lu <yinghai@meta.com>
|
2026-05-18 15:59:33 -07:00 |
|
|
Baizhou Zhang
|
6f892047ec
|
[misc] Throw error when single batch overlap is enabled on Hopper (#25509)
|
2026-05-18 14:51:32 -07:00 |
|
|
Liangsheng Yin
|
9e3bb9a307
|
[Spec] fold can_run_cuda_graph into EagleVerifyOutput; drop dead extend-after-decode check (#25566)
|
2026-05-18 14:24:54 -07:00 |
|