Edit model card

mlx-community/pixtral-12b-bf16

This model was converted to MLX format from mistral-community/pixtral-12b using mlx-vlm version 0.0.15. Refer to the original model card for more details on the model.

Use with mlx

pip install -U mlx-vlm
python -m mlx_vlm.generate --model mlx-community/pixtral-12b-bf16 --max-tokens 100 --temp 0.0
Downloads last month
381
Safetensors
Model size
12.7B params
Tensor type
BF16
·
Inference Examples
Inference API (serverless) does not yet support transformers models for this pipeline type.