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

Add MMMU benchmark results (#4491)

Browse Source 
Co-authored-by: Ravi Theja Desetty <ravitheja@Ravis-MacBook-Pro.local>
This commit is contained in:
Ravi Theja
2025-04-25 12:53:53 +05:30
committed by GitHub
parent b5be56944b
commit 7d9679b74d
4 changed files with 189 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
benchmark/mmmu/README.md
View File

@@ -21,10 +21,3 @@ It's recommended to reduce the memory usage by appending something ike `--mem-fr
```
python benchmark/mmmu/bench_hf.py --model-path Qwen/Qwen2-VL-7B-Instruct
```
Some popular model results:
1. Qwen/Qwen2-VL-2B-Instruct: 0.241
2. Qwen/Qwen2-VL-7B-Instruct: 0.255
3. Qwen/Qwen2.5-VL-3B-Instruct: 0.245
4. Qwen/Qwen2.5-VL-7B-Instruct: 0.242