Image Feature Extraction
Transformers
JAX
Safetensors
MLX
PyTorch
aimv2_vision_model
vision
custom_code
Eval Results (legacy)
Instructions to use apple/aimv2-3B-patch14-336 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use apple/aimv2-3B-patch14-336 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-feature-extraction", model="apple/aimv2-3B-patch14-336", trust_remote_code=True)# Load model directly from transformers import AutoImageProcessor, AutoModel processor = AutoImageProcessor.from_pretrained("apple/aimv2-3B-patch14-336", trust_remote_code=True) model = AutoModel.from_pretrained("apple/aimv2-3B-patch14-336", trust_remote_code=True) - MLX
How to use apple/aimv2-3B-patch14-336 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir aimv2-3B-patch14-336 apple/aimv2-3B-patch14-336
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
| { | |
| "architectures": [ | |
| "AIMv2Model" | |
| ], | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_aimv2.AIMv2Config", | |
| "AutoModel": "modeling_aimv2.AIMv2Model", | |
| "FlaxAutoModel": "modeling_flax_aimv2.FlaxAIMv2Model" | |
| }, | |
| "hidden_size": 3072, | |
| "image_size": 336, | |
| "intermediate_size": 8192, | |
| "model_type": "aimv2", | |
| "num_attention_heads": 24, | |
| "num_channels": 3, | |
| "num_hidden_layers": 24, | |
| "patch_size": 14, | |
| "projection_dropout": 0.0, | |
| "qkv_bias": false, | |
| "rms_norm_eps": 1e-05, | |
| "torch_dtype": "float32", | |
| "transformers_version": "4.46.3", | |
| "use_bias": false | |
| } |