mradermacher

/

model_requests

like 68

Let's quant it then :) Unfortunately, hf severely rate limited me at the moment, so while the model should be done quickly, I don't know when I can upload it (but probably within a few hours). You can watch how it progresses at http://hf.tst.eu/status.html - "repo fetch" usually means we can't create the repository.

Unfortunately, the config.json doesn't list its architecture (no architectures key - and it looks suspiciously empty), so transformers/llama.cpp can't detect what type of model it is.

Also, I had no idea chrome with an llm. Fascinating provenance.

Yeah, that doesn't look like an incomplete config file:

Yes, this is to be expected. You cannot run this as of now as this is just the model files, the code to run the model doesn’t exist yet.