https://huggingface.co/aquif-ai/aquif-3.5-Max-1205
yep, new model. this time thinking-only
bruh why did it post twice?? lol, sorry for that
bruh why did it post twice?? lol, sorry for that
Do not click "Create Discussion" two times, it doesn't stick, you just need to wait a few seconds to let it publish your comment due to a delay
It's queued!
You can check for progress at http://hf.tst.eu/status.html or regularly check the model
summary page at https://hf.tst.eu/model#aquif-3.5-Max-1205-GGUF for quants to appear.
The thread is a duplicate of >1582
Thanks for letting us know.
bruh why did it post twice?? lol, sorry for that
No problem. In worst case our system will tell us if I try to queue a model that was already queued in the past.
It's really nice to see an improved version of aquif-3.5-Max-42B-A3B with this model. 42B is a really convenient size as it is the largest many users can run on GPU at a decent quantization level. For example one could cheaply run this on 2x Intel Arc A770 or 1x RTX 5090 at a great quantization level with enough memory to spare for context.
thanks! yeah, 42B-A3B is a lot more accessible on consumer hardware than 235B-A22B or even GLM-Air-sized models