noctrex
/

Qwen3.5-122B-A10B-MXFP4_MOE-GGUF

Image-Text-to-Text

Model card Files Files and versions

Kind request for Qwen3.5-397B-A17B MXFP4 BF16

#2

by dehnhaide - opened Feb 26

Feb 26

Hi Noctrex,

First if all thanks for the hard work you put in releasing so many super quants! 🙏

As the title says, dare I kindly as for a MXFP4 / BF16 GGUF of Qwen3.5-397B-A17B.
I am on Ampere, 8x RTX 3090, and based on previous releases of your MXFP4 I would really like to test this version of quantization. The model seems highly capable and MXFP4 has been so far quite speedy on my configuration!

Many thanks in advance of the model gets released! ✌️

Owner Feb 26

Yeah, that's quite a large model. Let me see if I can find some space to quantize it.

Feb 27

Thanks for the MXFP4 version, it runs very well. An ablit. version would be nice 😄

Owner Feb 27

Here you go: https://huggingface.co/noctrex/Qwen3.5-397B-A17B-MXFP4_MOE-GGUF

Owner Feb 27

@matixxx Will quantize some huihui ablit versions soon

Feb 27

Here you go: https://huggingface.co/noctrex/Qwen3.5-397B-A17B-MXFP4_MOE-GGUF

Thank you for your help! Vielen Dank! 😎

Owner Feb 27

Kein Problem

Feb 27

@matixxx Will quantize some huihui ablit versions soon

na da bin ich schonmal gespannt drauf 😁

noctrex changed discussion status to closed 23 days ago

Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.

Tap or paste here to upload images

· Sign up or log in to comment