mlx-community

/

DeepSeek-R1-Distill-Llama-8B-4bit-mlx

like 1

Follow

MLX Community 12.8k

rope_scaling must contain keys {'factor', 'type'}
was facing this error when I tried to quantize the unquantized model. any inputs on why i am facing this and possible solutions

ADDITIONAL INFO:
Device: M1 pro, 32 gb ram
MLX lora training example script