What to do if ollama downloads big model and gets error resetting connection?

Question

enter image description here I'm using macos. When I ran ollama run llama3 in the terminal and got the following error：

pulling manifest 
Error: pull model manifest: Get "https://ollama.com/token?nonce=8UU-FvbYyGTesdUePHqRqQ&scope=repository%!A(MISSING)library%!F(MISSING)llama3%!A(MISSING)pull&service=ollama.com&ts=1714116326": read tcp 172.17.12.43:50708->34.120.132.20:443: read: connection reset by peer

How do I fix it?

I tried turning off the firewall and it didn't work too.

Answer 1

The issue pulling the model manifest usually happens due to CDN issues/configurations. Try using a VPN from different territory, i.e. USA and you may observe an improvement.

Answer 2

Update Ollama and this issue will be resolved