Thank you and a request!
#2
by
MidnightPhreaker
- opened
Thank you Fireworks!!! Good to have more NVFP4 models out there!
Would you mind giving cerebras/GLM-4.5-Air-REAP-82B-A12B the NVFP4 treatment?
Have an awesome day!
Just finished running it. Give it a try and see if it works for you? My quantization script seemed to complete just fine on it but I spent a while battling VLLM trying to get it to run and I'm clearly missing something in the environment to make it happy.
https://huggingface.co/Firworks/GLM-4.5-Air-REAP-82B-A12B-nvfp4
Absolute, bloody legend! Thank you!! I'll give it a crack right now and let you know, and again, thank you ! <3