Files
sglang/benchmark/long_json_decode
Lianmin Zheng 22085081bb release initial code
Co-authored-by: Ying Sheng <sqy1415@gmail.com>
Co-authored-by: Liangsheng Yin <hnyls2002@gmail.com>
Co-authored-by: Zhiqiang Xie <xiezhq@stanford.edu>
Co-authored-by: parasol-aser <3848358+parasol-aser@users.noreply.github.com>
Co-authored-by: LiviaSun <33578456+ChuyueSun@users.noreply.github.com>
Co-authored-by: Cody Yu <hao.yu.cody@gmail.com>
2024-01-08 04:37:50 +00:00
..
2024-01-08 04:37:50 +00:00
2024-01-08 04:37:50 +00:00
2024-01-08 04:37:50 +00:00
2024-01-08 04:37:50 +00:00

Run benchmark

Benchmark sglang

python3 -m sglang.launch_server --model-path codellama/CodeLlama-7b-instruct-hf --port 30000
python3 bench_sglang.py --num-questions 5 --parallel 1

Benchmark vllm

python3 -m vllm.entrypoints.api_server --tokenizer-mode auto --model codellama/CodeLlama-7b-instruct-hf  --disable-log-requests --port 21000 --gpu 0.97
python3 bench_other.py --backend vllm --num-questions 5

Benchmark guidance

python3 bench_other.py --backend guidance --num-questions 5 --parallel 1

Build dataset

pip install wikipedia
python3 build_dataset.py