Image Feature Extraction
Transformers
JAX
Safetensors
MLX
PyTorch
aimv2_vision_model
vision
custom_code
Eval Results (legacy)
Instructions to use apple/aimv2-large-patch14-336 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use apple/aimv2-large-patch14-336 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-feature-extraction", model="apple/aimv2-large-patch14-336", trust_remote_code=True)# Load model directly from transformers import AutoImageProcessor, AutoModel processor = AutoImageProcessor.from_pretrained("apple/aimv2-large-patch14-336", trust_remote_code=True) model = AutoModel.from_pretrained("apple/aimv2-large-patch14-336", trust_remote_code=True) - MLX
How to use apple/aimv2-large-patch14-336 with MLX:
# Download the model from the Hub pip install huggingface_hub[hf_xet] huggingface-cli download --local-dir aimv2-large-patch14-336 apple/aimv2-large-patch14-336
- Notebooks
- Google Colab
- Kaggle
- Local Apps
- LM Studio
Addition of HF compatible checkpoints
#2
by yaswanthgali - opened
First of all great work and thanks for these strong models. I'm working on the PR to add native support for AimV2 in Transformers library. It would be helpful to transfer the HF-compatible checkpoints to the official repo to make user's life easier. I've converted four checkpoints covering different types for starters.
PR: https://github.com/huggingface/transformers/pull/36625
- yaswanthgali/aimv2-large-patch14-336-HF
- yaswanthgali/aimv2-large-patch14-224-HF
- yaswanthgali/aimv2-large-patch14-224-lit-HF
- yaswanthgali/aimv2-large-patch14-native-HF
LMK if you want to update the existing checkpoints or a separate set of checkpoints for HF compatible ones which IMO is not required as previously we use to load current ones with trust_remote_code=True arg .
Closing this as completed, thx!
michalk8 changed discussion status to closed
look forward to huhu