Downloads · 30 days
50
13% of all-time downloads
qwen_4b_eagle3
Tengyunw/qwen_4b_eagle3
qwen_4b_eagle3 is a machine learning model from Tengyunw. Use it for the machine learning task on the model card, and read the license before you ship it in a product.
How to use To use Eagle3 with SGLang, first replace the qwen3.py file in SGLang’s directory (sglang/python/sglang/srt/models/) with the qwen3.py file from this project.
All-time downloads
372
Public
Parameters
218M
437 MB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors437 MB · 100%
Parameter types
How the weights are stored.
BF16218M · 100%
Model card
From the Hugging Face model README
How to use To use Eagle3 with SGLang, first replace the qwen3.py file in SGLang’s directory (sglang/python/sglang/srt/models/) with the qwen3.py file from this project.
The launch command for using Eagle3 with SGLang is:
python3 -m sglang.launch_server --model Qwen/Qwen3-4B-Instruct-2507 --speculative-algorithm EAGLE3 --speculative-draft-model-path Tengyunw/qwen3_4b_eagle3 --speculative-num-steps 6 --speculative-eagle-topk 10 --speculative-num-draft-tokens 32 --mem-fraction 0.9 --cuda-graph-max-bs 2 --dtype bfloat16
- Type
- llama
- Created
- Nov 27, 2025
- Updated
- Nov 27, 2025