GitLab

  - 在 C++ 算子中移除 query/key/residual 的 contiguous 强校验
  - 对非连续输入使用工作张量执行计算（优化路径与 fallback 路径统一）
  - 计算完成后将结果 copy_ 回原张量，保持 in-place 语义
  - 移除 qwen3 / qwen3_moe Python 前向中的 q、k.contiguous() 预处理

  launch_opt_rms_rope 增加 rot_dim=64/128 分发路径
  放宽 supports_qwen3_opt 条件，允许 rot_dim=64 进入优化分支
  qwen3 / qwen3_moe 将 q_bias、k_bias 参数统一为 q_residual、k_residual
  qwen3_moe 的自定义 op 注册增加重复注册保护"`

  - 在 vLLM _C 扩展中新增 rms_rotary_embedding_fuse（注册 op + CUDA kernel），减少对 LightOp 的硬依赖
  - 新增环境变量 VLLM_FUSED_RMS_ROPE_BACKEND=auto|vllm|lightop，auto 优先走 vLLM，缺失时回退 LightOp
  - 更新 Qwen3 / Qwen3-MoE 的 fused 路径按后端选择执行
  - 补充 tc_opt benchmark 结果解析脚本 benchmarks/tc_opt/test/parse_bench_results.py

实现了用于优化张量计算的 rms_mrope_fuse 和 rms_mrope_fuse_fake 方法
更新了 forward：在满足条件时走新的 M-RoPE 融合路径
增强了 Qwen3MoeModel 对动态参数维度的支持，以适配该功能

新增环境变量 VLLM_USE_MARLIN_W16A16_MOE，用于显式启用 Marlin W16A16 MoE experts
在 fused_moe 中当开关开启且实现可用时，调用 fused_experts_impl_w16a16_marlin
增加 Marlin W16A16 MoE 实现与 reduce 路径

set default_max_num_batched_tokens = 10240
update qwen3_moe of layernorm
off lightop of moe_fused_gate

Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: simon-mo <simon.mo@hey.com>

Signed-off-by: Tyler Michael Smith <tlrmchlsmth@gmail.com>
Signed-off-by: simon-mo <simon.mo@hey.com>

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

Signed-off-by: Woosuk Kwon <woosuk@thinkingmachines.ai>

Signed-off-by: Bill Nell <bnell@redhat.com>

Signed-off-by: whx-sjtu <2952154980@qq.com>

Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Huang Jie <92386084+JJJYmmm@users.noreply.github.com>
Co-authored-by: 松灵 <26085463+wulipc@users.noreply.github.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

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

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

[BUGFIX] GPTQ quantization compatibility for Qwen3 MOE models (AutoGPTQ and AutoRound-GPTQ) (#23994)
Signed-off-by: JartX <sagformas@epdcenter.es>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>

Signed-off-by: Lukas Geiger <lukas.geiger94@gmail.com>

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

Signed-off-by: rongfu.leng <rongfu.leng@daocloud.io>
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
Signed-off-by: rongfu.leng <lenronfu@gmail.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>

Signed-off-by: Yizhou Liu <liu_yizhou@outlook.com>

Signed-off-by: Gh0u1L5 <Gh0u1L5@outlook.com>

Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>

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

Signed-off-by: Zhiyu Cheng <zhiyuc@nvidia.com>

Signed-off-by: Tao He <linzhu.ht@alibaba-inc.com>

Signed-off-by: aladerran <aladerran@gmail.com>

Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>

Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>

Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>

Signed-off-by: Bill Nell <bnell@redhat.com>

Signed-off-by: noiji <>

Co-authored-by: xinnan.hou <hxn02029096@alibaba-inc.com>