【Hackathon 9th No.93】【RFC】为 Fastdeploy 新增 MiniMax-M1 模型 #1156
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Collaborator
|
|
||
| **核心技术路径**: | ||
| 1. **复用**: 最大化复用 GLM-4.5 PR 中已有的 Partial RoPE 和标准 GQA Attention 组件。 | ||
| 2. **翻译与开发**: 将 vLLM 的 `lightning_attn.py` (Triton) 翻译为高性能的 CUDA C++ 算子，以支持 MiniMax-M1 的线性注意力层。 |
There was a problem hiding this comment.
可以先用triton算子快速验证，同步开发高性能cuda kernel，FD目前支持使用triton算子
|
|
||
| --- | ||
|
|
||
| ### **Phase 1: [核心开发] 实现 Mamba/线性注意力 CUDA 算子 (2-4 周)** |
There was a problem hiding this comment.
目前看主要开发工作在线性注意力，可以先尝试接入triton算子快速验证下，如果attention性能不佳可以尝试实现cuda kenrel优化端到端性能。
chang-wenbin
approved these changes
Sep 19, 2025
luotao1
approved these changes
Sep 19, 2025
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
本文档为新增 MiniMax-M1 模型的 RFC，规划了从 CUDA 算子开发到模型整体集成的技术方案。