Compare

Line up your shortlist and see the differences at a glance.

Feature comparison of selected websites
	llama.cpp Run LLMs anywhere, in plain C/C++.	Ollama Run LLMs on your own machine.	Jan Open-source ChatGPT that runs offline.
Rating	—	—	—
Pricing	Free	Free	Free
Open source
Category	Local Inference	Local Inference	Local Inference
Key features	Runs quantized GGUF models CPU and GPU backends Powers Ollama, Jan, and more	One-command model running OpenAI-compatible local API Model library with quantizations	Local chat UI OpenAI-compatible local server Model hub built in
Pros	Runs on almost any hardware The performance reference	Total privacy, no cloud Dead-simple developer experience	Private, offline, and free Friendlier than raw llama.cpp
Cons	Command-line and low-level You manage models yourself	Needs decent RAM/GPU Local models trail frontier quality	Needs decent RAM Younger than LM Studio
	Visit github.com	Visit ollama.com	Visit jan.ai

Add a site to compare