Spaces:

huggingchat

/

chat-ui

like 1.1k

Running

When interacting with the OpenAssistant/oasst-sft-6-llama-30b model in the chat, I receive the following error in the top right of the screen:

Model OpenAssistant/oasst-sft-6-llama-30b does not exist

When I click on retry n times, then the conversation continues at some point.

I guess this happens because of too many users using the API, as the error comes from https://api-inference.huggingface.co/models/?

If my assumption is true, would it make sense to introduce some kind of auto-retry + timeout in chat-ui? Would love to contribute this!