- 24 Feb, 2026 1 commit
-
-
laibao authored
新增 router_capture 工具,用于按 num_tokens/rank 过滤并落盘 MoE router logits 在 Qwen3MoeSparseMoeBlock 中接入采集调用,并在 torch.compile 场景下自动跳过 补充 VLLM_MOE_ROUTER_CAPTURE* 环境变量
-
- 19 Feb, 2026 1 commit
-
-
王敏 authored
-
- 16 Feb, 2026 2 commits
- 11 Feb, 2026 3 commits
- 10 Feb, 2026 4 commits
- 09 Feb, 2026 1 commit
-
-
jujl1 authored
-
- 08 Feb, 2026 2 commits
- 06 Feb, 2026 6 commits
- 05 Feb, 2026 2 commits
- 04 Feb, 2026 8 commits
-
-
zhuwenwen authored
-
zhuwenwen authored
-
zhuwenwen authored
-
zhuwenwen authored
-
zhuwenwen authored
-
zhuwenwen authored
-
Michael Goin authored
Signed-off-by:Robert Shaw <rshaw@neuralmagic.com>
-
Michael Goin authored
[Bugfix] Disable RoutingMethodType.[Renormalize,RenormalizeNaive] TRTLLM per-tensor FP8 MoE (#33620) Signed-off-by:
mgoin <mgoin64@gmail.com> (cherry picked from commit e346e2d0 ) Signed-off-by:
Robert Shaw <rshaw@neuralmagic.com>
-
- 03 Feb, 2026 4 commits
-
-
zhuwenwen authored
-
zhuwenwen authored
-
Richard Zou authored
Signed-off-by:
Richard Zou <zou3519@gmail.com> (cherry picked from commit d9aa39a3)
-
Zhewen Li authored
Signed-off-by:
zhewenli <zhewen@inferact.ai> Co-authored-by:
zhewenli <zhewen@inferact.ai>
-
- 02 Feb, 2026 4 commits
-
-
csy0225 authored
Signed-off-by:
Jee Jee Li <pandaleefree@gmail.com> Co-authored-by:
i-zhangmingming <i-zhangmingming@stepfun.com> Co-authored-by:
xiewuxun <xiewuxun@stepfun.com> Co-authored-by:
zetaohong <i-hongzetao@stepfun.com> Co-authored-by:
Jee Jee Li <pandaleefree@gmail.com> (cherry picked from commit c3b40dc3)
-
René Honig authored
Signed-off-by:
mgoin <mgoin64@gmail.com> Co-authored-by:
mgoin <mgoin64@gmail.com> (cherry picked from commit 07978117)
-
Gregory Shtrasberg authored
Signed-off-by:
Gregory Shtrasberg <Gregory.Shtrasberg@amd.com> (cherry picked from commit 31aedfe7)
-
Michael Goin authored
Signed-off-by:
mgoin <mgoin64@gmail.com> (cherry picked from commit bfb9bdaf)
-
- 29 Jan, 2026 1 commit
-
-
zhuwenwen authored
not supported FlashMLASchedMeta
-
- 28 Jan, 2026 1 commit
-
-
zhuwenwen authored
-