llama-buddy
CLI wrapper to manage, download, and serve local LLMs via llama-server.
Tested on Linux
Use it
Linux x86_64
You need first
- Runtimes:
python>=3.10 - System packages:
llama-server
uv tool install llama-buddy
uv tool install llama-buddy
Check that it works
llb status 2>&1The output should contain “llama-server is not running.”.
First thing to try
llb status 2>&1Run llb status to see the current state of your local models and server.
What it does
- Manage, download, and serve local LLMsNot shown working
- Run llama-server in the backgroundNot tested
- Search HuggingFace and download modelsNot shown working
- Chat with a model interactivelyNot shown working
How far pickfit got
Linux x86_64
- Installs
- Starts
- 0 of 4 features work
- Works offline
Good for
- download a GGUF model from HuggingFace
- start llama-server in the background
- chat with a model via llama-cli
- inspect GGUF model metadata
- list local models with VRAM usage
An alternative to ollama.
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 31 Oct 2026
- Version 0.1.16
Differences the test machine hit
- README: Prerequisite: llama.cpp installed and llama-server on PATH
In the test: install succeeds without llama-server, but llb start fails with 'llama-server not found. Please install llama.cpp.'; wrapper works but serve/chat need binary - README: llb models shows table of models
In the test: without running server exits 1 with 'Could not connect to llama-server. Is it running?' rather than empty table