NVIDIA Ships Qwen 3.8 27B NVFP4 Quantized Model on Hugging Face

TheZachMueller · x · 2026-09-09

NVIDIA has released an NVFP4 quantized version of Qwen 3.8 27B on Hugging Face (nvidia/Qwen3.8-27B-NVFP4), as spotted and shared by TheZachMueller on Twitter.

The model uses the NVFP4 4-bit format optimized for inference on NVIDIA hardware. Its model card includes a chat template supporting image and video inputs, indicating it is a quantized release of a multimodal model, aimed at developers who want to run Qwen efficiently at low precision on NVIDIA GPUs.

Related event: NVIDIA releases official NVFP4 quantized Qwen3.8-27B(2 posts)→

Original post →

More from Models

Models channel →