AxionML/Qwen3.5-4B-NVFP4 · Only producing garbage in H200, cu130 with CUDA 13.0

Only producing garbage in H200, cu130 with CUDA 13.0

by Dsturb - opened 4 days ago

Hi,

I can not get any reasonable output from your quntisation. The Qwen 3 2507 is running well in NVFP4 even there is no native support.

In a similar position, just get gibberish output from this quant

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