LLM Extraction Stuck on “Processing” with LLaMA3.18BInstructQ5 Model, No Errors

[banyous]

Hi, and thanks so much for your hard work on this project!

I'm running the LLM information extraction using the LLaMA3.18BInstructQ5 model. The process starts and the screen shows “processing,” but the status never updates to finished (screenshot attached). I’m not seeing any errors in debug mode.

Has anyone encountered this issue or have suggestions on what might be going wrong?

[FWao]

What setup are you using here? A docker image or manual installation? Is the model a local one? What are the last things printed in the terminal / docker logs?

[ben-peter]

I'm having the same problem. I'm using Meta-Llama-3-8B-Instruct-Q5_K_M. I tried it with Docker. With your image and with one I built myself. The model is integrated locally. In the logs, it looks like the LLM isn't getting any jobs. It works with manual installation and local API.
The logs don't give any clues as to the problem.
A friend of mine uses an old version of LLMAIx and it works there.
I am happy to answer any further questions.