60 lines
1.2 KiB
Markdown
60 lines
1.2 KiB
Markdown
## Download data
|
|
```
|
|
bash download_data.sh
|
|
```
|
|
|
|
## Run benchmark
|
|
|
|
### Benchmark sglang
|
|
```
|
|
python -m sglang.launch_server --model-path meta-llama/Llama-2-7b-chat-hf --port 30000
|
|
```
|
|
|
|
```
|
|
python3 bench_sglang.py --nsub 10
|
|
```
|
|
|
|
```
|
|
# OpenAI models
|
|
python3 bench_sglang.py --backend gpt-3.5-turbo --parallel 8
|
|
```
|
|
|
|
### Benchmark vllm
|
|
```
|
|
python3 -m vllm.entrypoints.api_server --tokenizer-mode auto --model meta-llama/Llama-2-7b-chat-hf --disable-log-requests --port 21000
|
|
```
|
|
|
|
```
|
|
python3 bench_other.py --nsub 10 --backend vllm
|
|
```
|
|
|
|
|
|
### Benchmark lightllm
|
|
```
|
|
# A10G
|
|
python -m lightllm.server.api_server --tokenizer_mode auto --model_dir ~/model_weights/llama-2-7b-chat-hf --max_total_token_num 16000 --port 22000
|
|
|
|
# V100
|
|
python -m lightllm.server.api_server --tokenizer_mode auto --model_dir ~/model_weights/llama-2-7b-chat-hf --max_total_token_num 4500 --port 22000
|
|
```
|
|
|
|
```
|
|
python3 bench_other.py --nsub 10 --backend lightllm
|
|
```
|
|
|
|
|
|
### Benchmark guidance
|
|
```
|
|
python3 bench_other.py --nsub 10 --backend guidance --parallel 1 --n-ctx 4096 --model-path path/to/gguf
|
|
```
|
|
|
|
|
|
### Benchmark lmql
|
|
```
|
|
CUDA_VISIBLE_DEVICES=0,1 lmql serve-model meta-llama/Llama-2-7b-chat-hf --cuda --port 23000
|
|
```
|
|
|
|
```
|
|
python3 bench_other.py --nsub 10 --backend lmql --parallel 2
|
|
```
|