mlx-community/Qwen3-8B-4bit
Quantization vs Qwen/Qwen3-8B. mlx-community/Qwen3-8B-4bit is a quantized version of the other (U32/BF16).
5 derivatives checked
- VertexAGI/prism-roleplay-1.5-small: Partial fine-tune
- VertexAGI/prism-roleplay-1-small: Partial fine-tune
- Fenomen-Alex/ukrainian-municipal-accountability-qwen3-8b: Partial fine-tune
- VertexAGI/codeinluau-1-small: Partial fine-tune
- VertexAGI/codeinswift-1-small: Partial fine-tune
3 other repos hold byte-for-byte identical weight files.