DeepSeek-V4-Flash-6bit-mlx

Description

DeepSeek-V4-Flash-6bit-mlx is an open-weight text-generation model with roughly 6B parameters published by lemuralabs on the Hugging Face Hub. It has 784 downloads. Runs with mlx.

Categories

• Productivity

Tags

• deepseek_v4
• deepseek-v4
• deepseek
• quantized
• apple-silicon
• mixture-of-experts
• Mixture of Experts
• affine

Capabilities

• Text generation
• Conversational
• Library: mlx
• 6B parameters