GitLab

[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

support prefix cache on kme
fix the error in test_moe caused by moe align not supporting 511 and 211
multi-modal switching to torch implementation on z100l&k100

fix the error in test_moe caused by moe align not supporting 511
multi-modal switching to torch implementation on z100l&k100

This reverts commit 20b6cf64.

Signed-off-by: Andy Xie <andy.xning@gmail.com>

Signed-off-by: yewentao256 <zhyanwentao@126.com>

Signed-off-by: luka <luka@neuralmagic.com>

Signed-off-by: Kyle Sayers <kylesayrs@gmail.com>

[Refactor] Create a function util and cache the results for `has_deepgemm`, `has_deepep`, `has_pplx` (#20187)
Signed-off-by: yewentao256 <zhyanwentao@126.com>

Signed-off-by: Boyuan Feng <boyuan@meta.com>

Signed-off-by: Isotr0py <2037008807@qq.com>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>

Signed-off-by: Ye (Charlotte) Qi <yeq@meta.com>

Signed-off-by: Benjamin Chislett <benjamin.chislett@centml.ai>

Signed-off-by: simon-mo <simon.mo@hey.com>

Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>