Let's help help help devs.
Target: <200ms. Most LLM CLI tools take 10+ seconds.
| tool | time | version | install | command | pr |
|---|---|---|---|---|---|
| vllm | 14263ms | 0.13.0+cpu | pip install vllm | vllm --help | - |
| sglang | TBD | - | - | - | - |
| ollama | 13ms | 0.13.5 | curl -fsSL https://ollama.com/install.sh | sh | ollama --help | - |
| llama.cpp | TBD | - | - | - | - |
| transformers | 8232ms | 4.57.3 | pip install transformers | transformers-cli --help | - |