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

Update ci test and doc for MTP api change (#5952)

Browse Source
This commit is contained in:
Ke Bao
2025-05-02 00:30:27 +08:00
committed by GitHub
parent de2faef97e
commit ebaba85655
6 changed files with 66 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
docs/references/deepseek.md
View File

@@ -153,12 +153,10 @@ The precompilation process typically takes around 10 minutes to complete.
**Description**: SGLang implements DeepSeek V3 Multi-Token Prediction (MTP) based on [EAGLE speculative decoding](https://docs.sglang.ai/backend/speculative_decoding.html#EAGLE-Decoding). With this optimization, the decoding speed can be improved by **1.8x** for batch size 1 and **1.5x** for batch size 32 respectively on H200 TP8 setting.
**Usage**:
Add arguments `--speculative-algorithm`, `--speculative-draft-model-path`,
`--speculative-num-steps`, `--speculative-eagle-topk` and `--speculative-num-draft-tokens` to enable this feature. For example:
Add arguments `--speculative-algorithm`, `--speculative-num-steps`, `--speculative-eagle-topk` and `--speculative-num-draft-tokens` to enable this feature. For example:
```
python3 -m sglang.launch_server --model-path deepseek-ai/DeepSeek-V3-0324 --speculative-algorithm EAGLE --speculative-draft-model-path lmsys/DeepSeek-V3-0324-NextN --speculative-num-steps 1 --speculative-eagle-topk 1 --speculative-num-draft-tokens 2 --trust-remote-code --tp 8
python3 -m sglang.launch_server --model-path deepseek-ai/DeepSeek-V3-0324 --speculative-algorithm EAGLE --speculative-num-steps 1 --speculative-eagle-topk 1 --speculative-num-draft-tokens 2 --trust-remote-code --tp 8
```
- The draft model are available at huggingface: [lmsys/DeepSeek-V3-0324-NextN](https://huggingface.co/lmsys/DeepSeek-V3-0324-NextN), [lmsys/DeepSeek-R1-NextN](https://huggingface.co/lmsys/DeepSeek-R1-NextN). It can also be exported from original DeepSeek-V3/R1 model with [export_deepseek_nextn.py](https://github.com/sgl-project/sglang/blob/main/scripts/export_deepseek_nextn.py) script.
- The best configuration for `--speculative-num-steps`, `--speculative-eagle-topk` and `--speculative-num-draft-tokens` can be searched with [bench_speculative.py](https://github.com/sgl-project/sglang/blob/main/scripts/playground/bench_speculative.py) script for given batch size. The minimum configuration is `--speculative-num-steps 1 --speculative-eagle-topk 1 --speculative-num-draft-tokens 2`, which can achieve speedup for larger batch sizes.
When using FlashInfer MLA wrapper (`--attention-backend flashinfer`) with speculative decoding, set the `--speculative-eagle-topk` parameter to `1`. The FlashAttention 3 backend also only supports `--speculative-eagle-topk 1`.
- To enable DeepSeek MTP for large batch sizes (>32), there are some parameters should be changed (Reference [this discussion](https://github.com/sgl-project/sglang/issues/4543#issuecomment-2737413756)):