MLX quantization of DeepSeek V4.1 Flash ships for Apple Silicon

The poster calls DeepSeek-V4.1-Flash-1017 the best local model and releases a high-quality, fast MLX quantization for Apple Silicon.

raullen

@RaullenArchived 10 days agoOriginally posted 2026.10.22

raullen @RaullenOriginally posted 2026.10.22

Open original on X

DeepSeek-V4.1-Flash-1017 is simply the best local model you can run 🔥 We built the highest-quality, fastest MLX quantization, available on Apple Silicon right now 😸

This post quotes another post: “rapid-mlx ships DeepSeek V4.1 Flash MLX quant for 256GB+ Macs”

x.com/raullen/status/2113027612767572055Archived 2026.10.22 11:58

Open original on X