Minachist
/

Qwen3.8-Flash-Next-INT4-Mixed-AutoRound

Image-Text-to-Text

compressed-tensors

Model card Files Files and versions

Resources

View closed (0)

How is this running for you? Looking for feedback from users

#3 opened about 2 months ago by

Patch Updates

#1 opened about 2 months ago by

~140 t/s decode on 2x RTX 5090

#7 opened 20 days ago by

Feedback: Qwen 3.8 Flash Next on 3X3090+MTP

#6 opened about 1 month ago by

bugreport: PLE n-gram table left all-zero on disk (correct size, st_blocks=0)

#5 opened about 1 month ago by

Victor-the-same

Feedback: Qwen3.8 Flash Next on 4× RTX 3090

#4 opened about 1 month ago by

Qwen3.8 Flash-Next 3×3090+MTP — CUDA illegal access / PP rank failure under two live streams

#2 opened about 2 months ago by