Image-Text-to-Text
NInfer
qwen3.8
nvfp4
w4a4
dflash2
speculative-decoding
blackwell
multimodal
conversational
cuda
rtx-5090
Eval Results (legacy)
Instructions to use cometkim/Qwen3.8-27B-nvfp4full-NInfer with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- NInfer
How to use cometkim/Qwen3.8-27B-nvfp4full-NInfer with NInfer:
# No code snippets available yet for this library. # To use this model, check the repository files and the library's documentation. # Want to help? PRs adding snippets are welcome at: # https://github.com/huggingface/huggingface.js
- Notebooks
- Google Colab
- Kaggle
Artifact v3: patch-less custom weights
#6
by cometkim - opened
Upstream NInfer engine introduced custom weights loading
So this artifact has been updated to use updated format recipe.
Now this profile is working on upstream engine without patches! (Only MTP, DFlash2 weights still need patch yet)