GPT4All-J-MXFP4-mlx

Introduce

Quant by mlx-lm 0.31.3 with new pipeline for GPT-J model_type

mlx_lm.generate mlx_lm.generate --model /path/to/model --prompt "What is different between CPU and GPU?" --max-tokens 2048

Safetensors

Model size

1B params

Tensor type

F32

U32

MLX

Hardware compatibility

4-bit

Base model

Quantized

(1)

this model