Struggling with getting the model to work after launching it. #5998
Replies: 2 comments
|
A 503 here means Odysseus is not getting a usable response from the configured model-serving endpoint, but the screenshot hides the underlying error. One important detail is that Please also share, with keys and passwords redacted: how you installed Odysseus; the Odysseus version/commit or exact image tag; Immediately after reproducing the failure, please include the last 200 lines from the Odysseus container and whichever service is actually hosting this model. If possible, open browser DevTools, select Network, open |
|
ERROR 503 happened to me too, I don't know why, It only worked when I turned on "Agent" mode and assigned a "Workspace" folder, only then it responded, and after that I could turn it back off into "Chat" mode. What was my question or prompt❓ I only said "Hello" in an empty new chat |
Uh oh!
There was an error while loading. Please reload this page.
Installed docker and launched ollama and odysseus in docker.
Installed 3 different low-mem models to test out, but when I launch a model, start a new chat, select the model in the top right corner of the chat, and send a message I just get 503 errors. (See example below)
Tried using Gordon in Docker to figure out if there was some kind of error between the odysseus container and the ollama container and gordon did a bunch of things, however I still cant get it to respond to any messages.
Setting up LMStudio was an absolute breeze compared to this, but I'd love to get this working and try it out.
Can anyone point me in the right direction? Thanks.
All reactions