llm --help

Let's help help help devs.

Target: <1000ms. Several important LLM CLI tools take multiple seconds. PRs welcome ❤️

Run: Jul 16, 2026 12:00 UTC / CPU: Core™ i7-8700K; GPU: 1x RTX 5060 Ti

tool cold warm version
tensorrt-llm --help30726ms10853ms1.2.1
vllm --help18643ms6870ms0.25.1
VLMEvalKit --help14993ms4480msv0.2
sglang --help14614ms4444ms0.5.9
transformers --help7657ms2166ms5.14.1
datasets --help2709ms694ms5.0.0
llm --help1451ms646ms0.31.1
openai --help959ms421ms2.34.0
hf --help844ms331ms1.23.0
lm-eval --help1396ms229ms0.4.12
langchain-cli --help684ms209ms0.0.37
tokenspeed --help477ms178ms0.1.0@f3d14a8
ollama --help12ms12ms0.32.0
llama.cpp --help15ms11msb10042