infai
Control plane TUI to run llama.cpp and vLLM models via launch profiles with telemetry.
Tested on Linux
Use it
Linux x86_64
Download v0.24.1 release tarball and install binary to ~/.local/bin
mkdir -p ~/.local/bin && curl -sL https://github.com/dipankardas011/infai/releases/download/v0.24.1/infai_0.24.1_linux_amd64.tar.gz -o /tmp/infai-clean.tar.gz && tar -xzf /tmp/infai-clean.tar.gz -C /tmp && install -m 0755 /tmp/infai ~/.local/bin/infai
Where the README and reality differ
- README: curl -sL .../install.sh | bash installs to /usr/local/bin via sudo
What worked: install to ~/.local/bin (e.g. install -m 0755 /tmp/infai ~/.local/bin/infai) as non-root user - README: install.sh resolves version from /releases/latest, which now returns infaiw-v0.30.0 so the constructed infai_* URL 404s
What worked: pin an infai v* tag (e.g. v0.24.1) and download infai_0.24.1_linux_amd64.tar.gz from that release
Check that it works
infai --versionThe output should contain “infai 0.24.1”.
First thing to try
infaiRun infai in a terminal to open the TUI.
What it does
These come from the README; pickfit didn’t check them one by one.
- Run and monitor many models from one screenNot tested
- Search Hugging Face Hub and download modelsNot tested
- Run agents as a background service with sessionsNot tested
- Hand off work to subagents through sidecar loopsNot tested
How far pickfit got
Linux x86_64
- Installs
- Starts
- Main function works
- Works offline
macOS: the project says it runs there; pickfit hasn’t tried it.
Good for
- run and monitor many local models from one TUI
- download GGUF models from Hugging Face
- monitor per-model CPU, RAM and VRAM live
- check whether a model fits hardware before launch
How pickfit tested it
Each run starts from a clean machine and follows the steps above. How pickfit verifies
Linux x86_64
Its main function works
- Replayed from scratch
- Works offline
- Checked on 28 Oct 2026
- Version 0.24.1