Resources

Quant requests?

#25 opened 27 days ago by

pathosethoslogos

run on DGX SPARK

#23 opened about 1 month ago by

Muyanghao

Updated files for UD3.0 final?

➕ 10

#21 opened about 1 month ago by

HumanTrim

Any chance of Unsloth's Qwen3.8 27B NVFP4 DSPARK/DFLASH2?

#20 opened about 1 month ago by

MaMo7x

Update config.json

#19 opened about 1 month ago by

dsikka

output quality, not yet there

🤯 5

#18 opened about 2 months ago by

Rieker

How to finetune this model?

#17 opened about 2 months ago by

antonydaubuoi

TP2 benchmark on 2x RTX 5070 Ti (16GB) -- vision works, decode speed + VRAM findings

👍 3

#16 opened about 2 months ago by

curiouspp8

Running with SGLang

#14 opened about 2 months ago by

verycoldwinter

Using this on vllm but excessive tbinking

👀➕ 3

#13 opened about 2 months ago by

crystech

Changelog?

#12 opened about 2 months ago by

sparse-simian

Qwen3.8-27B Serving Configs: DGX Spark vLLM NVFP4

🤗🚀 10

#7 opened about 2 months ago by

erdal

Why no NVFP4 GGUF?

👍👀 10

#6 opened about 2 months ago by

mattzink

What is the target size for NVFP4?

#5 opened about 2 months ago by

musa334

Can it be reduced much further?

👍 1

#4 opened about 2 months ago by

Duonglv

So fast and great. Thank so muchhhh.

#3 opened about 2 months ago by

Duonglv

Thank you

#1 opened about 2 months ago by

pubdddq23erq