mradermacher

/

model_requests

like 67

That's already been tried, but before the time where I had detailed logs of why it failed. Let's try again :)

Looks fine, imatrix generated fine, this usually means it will succeed. In a few hours it will be done, or you can check the status display at http://hf.tst.eu/status.html

Oops, I think it quantized the older version without v1.
The requested "version is an improvement on the v0. Use this 8b for best results."
https://huggingface.co/NaniDAO/Meta-Llama-3.1-8B-Instruct-ablated-v1
Thanks!

Don't worry, your original url was correct and it quantized the correct model. You can go to the model page and see the available quantizartions in the lower right. As of this writing, it only finished the static ones, but the imatrix ones should be there soon (unfortunately it was allocated to the slowest node).

Thanks so much for the speedy responses! :)