DeepSeek-V4-Flash-Vision-Exp appears on HF with unsloth GGUF quants

unsloth · hf · 2026-09-04

DeepSeek-V4-Flash-Vision-Exp, an experimental vision-capable model from DeepSeek, is trending on Hugging Face via unsloth's GGUF quantization. The repo is MIT-licensed, imatrix-quantized, and transformers-compatible, enabling local deployment — hinting that DeepSeek is testing a new lightweight multimodal Flash model.

Related event: DeepSeek V4 Flash Vision Exp Merges Vision Support, GGUF Quants Available(2 posts)→

Original post →

More from Models

Models channel →