Tencent / Tencent/AngelSlim

Unable to replicate the performance of Qwen3 A3B eagle3

Open
#9 2 comments 0 reactions 1 assignee View on GitHub

@liusong1222 is already working on this.

Since Jul 15, 2025.

Dominant language
Python
Stars
1.7k
Forks
181
Avg merge
1d 3h
Merged PRs (30d)
2

Description

Poor performance with Qwen3 A3B eagle3.

Without eagle3: ~130 tokens/s
With eagle3: ~60 tokens/s

sglang version: 0.4.9.post2

Launch commands:

python3 -m sglang.launch_server \
--model-path Qwen--Qwen3-30B-A3B \
--reasoning-parser qwen3 --tool-call-parser qwen25 --attention-backend fa3 \
--speculative-algorithm EAGLE3 --speculative-draft-model-path AngelSlim--Qwen3-a3B_eagle3 \
--speculative-num-steps 6 --speculative-eagle-topk 10 --speculative-num-draft-tokens 32

Need your help. Thanks!

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.