Any-to-Any
Transformers
Safetensors
MLX
minicpmo
image-feature-extraction
minicpm-o
minicpm-v
multimodal
full-duplex
custom_code
4-bit precision
Instructions to use mlx-community/MiniCPM-o-4_5-4bit with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use mlx-community/MiniCPM-o-4_5-4bit with Transformers:
# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("mlx-community/MiniCPM-o-4_5-4bit", trust_remote_code=True, dtype="auto") - MLX
How to use mlx-community/MiniCPM-o-4_5-4bit with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir MiniCPM-o-4_5-4bit mlx-community/MiniCPM-o-4_5-4bit
- Notebooks
- Google Colab
- Kaggle
- Local Apps Settings
- LM Studio
| { | |
| "image_processor_type": "MiniCPMVImageProcessor", | |
| "feature_extractor_type": "MiniCPMAAudioProcessor", | |
| "auto_map": { | |
| "AutoProcessor": "processing_minicpmo.MiniCPMOProcessor", | |
| "AutoImageProcessor": "processing_minicpmo.MiniCPMVImageProcessor", | |
| "AutoFeatureExtractor": "processing_minicpmo.MiniCPMAAudioProcessor" | |
| }, | |
| "processor_class": "MiniCPMOProcessor", | |
| "max_slice_nums": 9, | |
| "scale_resolution": 448, | |
| "patch_size": 14, | |
| "use_image_id": true, | |
| "image_feature_size": 64, | |
| "im_start": "<image>", | |
| "im_end": "</image>", | |
| "slice_start": "<slice>", | |
| "slice_end": "</slice>", | |
| "unk": "<unk>", | |
| "im_id_start": "<image_id>", | |
| "im_id_end": "</image_id>", | |
| "slice_mode": true, | |
| "audio_pool_step": 5, | |
| "norm_mean": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "norm_std": [ | |
| 0.5, | |
| 0.5, | |
| 0.5 | |
| ], | |
| "version": 4.5 | |
| } |