vllm-project/vllm

[Feature]: Use kernel abstraction for fp4 scaled mm

Open

#31,823 建立於 2026年1月6日

在 GitHub 查看
 (12 留言) (0 反應) (1 負責人)Python (80,034 star) (16,816 fork)batch import
feature requesthelp wantedunstale

描述

🚀 The feature, motivation and pitch

See MPLinearKernel and ScaledMMLinearKernel.

Alternatives

No response

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.

貢獻者指南

技術棧
python
領域
backendmachine learningperformance
議題類型
feature
難度面向新貢獻者的預計實作難度，1 表示很小改動，5 表示專家級工作。
4
預計時間有經驗貢獻者完成調查、實作、測試並準備 pull request 的粗略時間範圍。
over 1 week
活動狀態議題目前的可參與程度：新鮮、活躍、陳舊、阻塞或等待維護者輸入。
blocked
清晰度議題是否清楚說明預期改動、驗收標準和下一步。
needs investigation
前置要求
Familiarity with vllm's linear kernel classesUnderstanding of fp4 quantizationCUDA kernel programming
新手友善度1-100 的估計分數，表示該議題對首次貢獻者的友善程度。
15
研究方向
Examine the existing MPLinearKernel and ScaledMMLinearKernel implementations in the vllm codebase to understand the current kernel abstraction for scaled matrix multiplication. Then design a new kernel abstraction specifically for fp4 (4 bit floating point) scaled MM, considering the discussion in the issue comments (12 comments) for any design constraints or preferences. No pull requests are linked, so this is a clean design task. Key files to review: likely in vllm/model executor/layers/ or similar paths.