https://huggingface.co/rhplus0831/maid-yuzu-v8-alter
Would you mine re-quanting rhplus0831/maid-yuzu-v8-alter, particularly the imatrix quants? It's not my model, but one I particularly enjoy. It's broken with the current version of llama.cpp.
I don't mind, as long as it has a user :) It's queued, and you can watch its progress at http://hf.tst.eu/status.html, or simply wait till it appears.
Thanks!
Was this one too broken to quant? :(
uh.... I am pretty sure I queued it, but I can't find the outcome in my logs, other than the first attempt in april.
I trust my submit log, so ... I clearly never submitted it after deleting the repositories. Sigh.
Anyway, it's definitely queued now, and the fastest node is working on it at highest priority :)
Everything looks fine. I have no other explanation than around that time, both my parents had to go to hospital and my main file server had catastrophic filesystem corruption and my life generally was in disarray (don't worry, parents are fine again and my server is still restoring from backup).
Still, it's very uncharacteristic of me to write "queued" without actually queuing...
Anyway, we have an imatrix, static quants, and likely soon have imatrix ones. Little can go wrong at this point. Finally...
Understandable, you had more important things on your mind at the time. I'm glad everyone is alright now. Thank you so much!