EngineX-Hygon/sglang
5
0
Fork 0
Code Issues Pull Requests Actions 7 Projects Releases Wiki Activity

[Revision] Add fast decode plan for flashinfer mla (#4012)

Browse Source
This commit is contained in:
Baizhou Zhang
2025-03-05 11:20:41 -08:00
committed by GitHub
parent 71ab0dabe0
commit fc91d08a8f
9 changed files with 145 additions and 34 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
docs/references/deepseek.md
View File

@@ -83,7 +83,7 @@ Please refer to [the example](https://github.com/sgl-project/sglang/tree/main/be
- **Weight Absorption**: By applying the associative law of matrix multiplication to reorder computation steps, this method balances computation and memory access and improves efficiency in the decoding phase.
- **Flashinfer MLA Wrapper**: By providing `--enable-flashinfer-mla` argument, the server will use MLA kernels customized by Flashinfer. More details can be referred to [this document](https://docs.flashinfer.ai/api/mla.html). (In Experiment Stage)
- **Flashinfer MLA Wrapper**: By providing `--enable-flashinfer-mla` argument, the server will use MLA kernels customized by Flashinfer. More details can be referred to [this document](https://docs.flashinfer.ai/api/mla.html). Under long input scenarios, flashinfer mla can improve performance significantly. Optimized triton kernels will be used when flashinfer mla is turned off.
- **FP8 Quantization**: W8A8 FP8 and KV Cache FP8 quantization enables efficient FP8 inference. Additionally, we have implemented Batched Matrix Multiplication (BMM) operator to facilitate FP8 inference in MLA with weight absorption.