Skip to content

Latest commit

 

History

History
23 lines (15 loc) · 435 Bytes

README.md

File metadata and controls

23 lines (15 loc) · 435 Bytes

DeServe Benchmark

Before running the benchmark, take a look at the source code of available options.

Download Datasets

git lfs install
git clone https://huggingface.co/datasets/anon8231489123/ShareGPT_Vicuna_unfiltered datasets/ShareGPT_Vicuna_unfiltered

Benchmarking Online vLLM

python -m src.benchmark.online_vllm

Benchmarking Offline vLLM

python -m src.benchmark.offline_vllm