llama3 ollama version errors

[Sjdavis25]

I've seen online that ollama works when using Gemma2 but nothing mething any ollama version of llama3. I want to create a chatbot that is based off of my local version of llama3 but every time I upload my ollama version of llama3 no matter what the embedded method is I can never recive a proper response. Whenever I try and use the playground chat I receive the message "There was an error processing your request" and whether or not I have data sources uploaded does not seem to affect this issue either.

[edwhiskey]

Did your problem solved? I tried Gemma2 and it doesn't work as well.

[Sjdavis25]

No, unfortunately I'm still having the same problem, still waiting for some sort of response

[n4ze3m]

Hey, sorry for the late reply. Can you share the logs, if possible?

[Sjdavis25]

[n4ze3m]

Are you using admin/api/chat or the playground (UI)

[Sjdavis25]

I'm using the playground UI

[edwhiskey]

my problem solved. You should pick the model which is uploaded not the one you named.
For me, I unseen all the unuse mode (e.g. from Open AI or Github), so I can get a clear list from Ollama.

[Sjdavis25]

When I try to pick a model I oly get the ones I named, am I uploading my models wrong?

[n4ze3m]

Hey, delete the current model, then go to Admin > Application, and turn on the fetch Ollama model dynamically

[Sjdavis25]

Got it to work thank you!!