search

lmstudio-ai / .github

34 stars 3 forks source link

Client Disconnect #10

Open nsteblay opened 5 months ago

nsteblay commented 5 months ago

2024-01-13 16:53:38.945] [INFO] [LM STUDIO SERVER] Client disconnected. Stopping generation... (if the model is busy processing the prompt, it will finish first)) [2024-01-13 16:53:39.884] [INFO] [LM STUDIO SERVER] Model is currently generating. Queuing request..

This happens consistently after processing several requests to the server. I need to quit LM Studio and restart after this happens.

I'm running a Python application using OpenAI on MacOS M1 with 64GB or memory.

yagil commented 5 months ago

Client disconnected means that the program making the API request to the server either exited or aborted the request in some way. So based on these bits of info you shared I'm inclined to say the issue is likely in the client code.

If you need help debugging your client code further you're welcome to post in the "projects" channel on the LM Studio discord: https://discord.gg/aPQfnNkxGC. Please share code, logs, and any other information you are comfortable uploading there.