frscrcc/Llama-3.3-70B-Instruct-abliterated-mlx-4Bit

Quantization vs huihui-ai/Llama-3.3-70B-Instruct-abliterated. frscrcc/Llama-3.3-70B-Instruct-abliterated-mlx-4Bit is a quantized version of the other (U32/F16).

On the Hub · Downloads on Model Pulse