mlx-community/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4-8-bit

Quantization vs Goekdeniz-Guelmez/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4. mlx-community/Josiefied-Qwen2.5-14B-Instruct-abliterated-v4-8-bit is a quantized version of the other (U32/F16).

On the Hub · Downloads on Model Pulse