Run large language models locally
A descrição completa deste programa ainda não foi traduzida, por isso aparece em inglês.
Ollama is a free tool that lets you download and run large language models (LLMs) — open AI chat models such as Llama, Mistral, Gemma and others — directly on your own computer. Everything runs locally, so your prompts and data stay on your PC and the models work without an internet connection once downloaded.
Ollama handles the technical work of downloading models and running them efficiently on your hardware. You can chat with a model from the command line or use the simple app, and developers can connect their own programs through Ollama's local API. Many third-party chat interfaces and coding tools also support Ollama.
Running AI models locally needs decent hardware: plenty of RAM and ideally a modern graphics card. Smaller models run on ordinary laptops but are less capable than large cloud-based assistants. For privacy-conscious users and developers experimenting with AI, Ollama is the easiest way to get started.
Prós
- Runs offline
- Easy model downloads
Contras
- Needs a strong PC