Jan
gui, server and inference engine all apache 2.0 — the only tool here where every layer is genuinely open.
the cleanest answer in the category since ollama's gui went closed — smaller and less polished, and the only one with nothing to explain.
gui, local openai-compatible server and the cortex inference engine underneath are all apache 2.0 in the same repository. of the nine tools here that is a claim only jan, llama.cpp, gpt4all and localai can make, and jan is the only one of those four that is also a finished desktop application.
practically it does what you want: cuda, metal and vulkan backends, gguf models, an openai-compatible endpoint so existing code points at it unchanged, and installers for macos, windows and linux.
the description above is ours, condensed from the ranking. pricing moves — check it on the vendor's own page before you rely on it.
- category
- local llm tools
- pricing
- free
- website
- github.com
the cleanest answer in the category since ollama's gui went closed — smaller and less polished, and the only one with nothing to explain.
we researched this category against vendors' own pricing pages and licence files. that is where this line comes from — not from the vendor, and not from anything they paid for.