llm-inference / library
MLC LLM
Capability: MLC LLM
Use it when
- compile language models for laptops, phones, GPUs, or browsers
- serve the same optimized model across heterogeneous devices
What it solves
Not the fit when
- model compilation and device support vary by platform
Install
Install the platform-specific `mlc-llm` package from the documented wheel index
Invoke
Compile or select a supported model, then run `mlc_llm serve MODEL` or use the Python `MLCEngine` API.
Alternatives
No reviewed alternatives recorded yet.