John Dee on Nostr: Why don't you have a self-hosted LLM yet? ...
Why don't you have a self-hosted LLM yet?
8 GB is enough to run llama3 8b, and larger models work, just slower. ollama with open-webui seems good so far.
https://github.com/ollama/ollama
https://github.com/open-webui/open-webui
8 GB is enough to run llama3 8b, and larger models work, just slower. ollama with open-webui seems good so far.
https://github.com/ollama/ollama
https://github.com/open-webui/open-webui