vllm-project/vllm

[Feature]: Fused Kernel for GPT-OSS Router

Open

#28,986 创建于 2025年11月19日

在 GitHub 查看
 (10 评论) (0 反应) (0 负责人)Python (80,034 star) (16,816 fork)batch import
feature requestgood first issuehelp wantedstale

描述

🚀 The feature, motivation and pitch

  • Right now, we spend ~3.5% of the layer in the expert selection
  • The operation is unfused

Write a fused kernel like we have for deepseek grouped_topk

Alternatives

  • torch compile
  • triton
  • cuda

Additional context

No response

Before submitting a new issue...

  • Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

贡献者指南

技术栈
pythonccpp
领域
backendperformance
议题类型
feature
难度面向新贡献者的预计实现难度，1 表示很小改动，5 表示专家级工作。
5
预计时间有经验贡献者完成调查、实现、测试并准备 pull request 的粗略时间范围。
over 1 week
活动状态议题当前的可参与程度：新鲜、活跃、陈旧、阻塞或等待维护者输入。
active
清晰度议题是否清楚说明期望改动、验收标准和下一步。
mostly clear
前置要求
GPU programmingCUDA/Triton kernel developmentvLLM architectureAttention mechanism
新手友好度1-100 的估计分数，表示该议题对首次贡献者的友好程度。
8
研究方向
Examine the existing deepseek grouped topk fused kernel in the vLLM repository (likely in csrc/ or kernels/). Look at the current GPT OSS router implementation (e.g., in vllm/model executor/layers/fused moe/ or similar). Identify the unfused operations and consider how to fuse them. Review issue comments for any initial design suggestions or performance profiling data. Refer to the alternatives mentioned (Triton vs CUDA) and evaluate trade offs. Consider writing a prototype in Triton for rapid iteration.