Files
sglang/benchmark/mmmu/README.md
2025-05-11 00:14:09 +08:00

26 lines
543 B
Markdown

## Run evaluation
### Evaluate sglang
Host the VLM:
```
python -m sglang.launch_server --model-path Qwen/Qwen2-VL-7B-Instruct --port 30000
```
It's recommended to reduce the memory usage by appending something like `--mem-fraction-static 0.6` to the command above.
Benchmark:
```
python benchmark/mmmu/bench_sglang.py --port 30000 --concurrency 16
```
You can adjust the `--concurrency` to control the number of concurrent OpenAI calls.
### Evaluate hf
```
python benchmark/mmmu/bench_hf.py --model-path Qwen/Qwen2-VL-7B-Instruct
```