unsloth Releases Quantized DeepSeek-V4-Flash-0731 Model

unsloth · hf · 2026-07-31

unsloth has released the DeepSeek-V4-Flash-0731 model on Hugging Face. Designed for text-generation pipelines and based on the deepseekv4 architecture, this release provides a quantized version of the base model. It is licensed under MIT and supports conversational use cases.

Related event: Unsloth Releases Quantized DeepSeek V4 Flash 0731 for Local Deployment(6 posts)→

Original post →

More from Models

Models channel →