- 10 Feb, 2026 1 commit
-
-
zhuwenwen authored
[opt] 优化epsp代码, 零消耗添加epsp update VLLM_USE_FUSED_RMS_ROPE=0 (default). for qwen3, VLLM_USE_FUSED_RMS_ROPE=1 (default) feat(moe/marlin): Marlin W16A16 MoE 自动探测并预打包(去掉手动开关) perf(qwen3): 融合 q/k RMSNorm + RoPE fused_moe_fp8接入lmslim
-
- 19 Aug, 2025 1 commit
-
-
zhuwenwen authored
-
- 24 Jul, 2025 1 commit
-
-
zhuwenwen authored
-
- 27 Jun, 2025 1 commit
-
-
Chendi.Xue authored
Signed-off-by:Chendi.Xue <chendi.xue@intel.com>
-
- 26 Jun, 2025 1 commit
-
-
Kunshang Ji authored
Signed-off-by:Kunshang Ji <kunshang.ji@intel.com>
-
- 03 Jun, 2025 1 commit
-
-
Simon Mo authored
Signed-off-by:simon-mo <simon.mo@hey.com>
-
- 28 Apr, 2025 1 commit
-
-
Lucas Wilkinson authored
Signed-off-by:
Lucas Wilkinson <lwilkinson@neuralmagic.com> Co-authored-by:
Jee Jee Li <pandaleefree@gmail.com> Co-authored-by:
Aaron Pham <contact@aarnphm.xyz>
-
- 23 Mar, 2025 1 commit
-
-
Lucas Wilkinson authored
Signed-off-by:
Lucas Wilkinson <lwilkinson@neuralmagic.com> Signed-off-by:
Lucas Wilkinson <lwilkins@redhat.com>
-
- 21 Mar, 2025 1 commit
-
-
Travis Johnson authored
Signed-off-by:Travis Johnson <tsjohnso@us.ibm.com>
-
- 20 Mar, 2025 1 commit
-
-
Mickaël Seznec authored
Signed-off-by:Mickael Seznec <mickael@mistral.ai>
-