Aimv2 Large Patch14 224

🚀 Transformers Library - AIMv2 Model

The AIMv2 family of vision models pre - trained with a multimodal autoregressive objective, offering high - performance on various vision tasks.

🚀 Quick Start

The transformers library provides easy - to - use interfaces for the AIMv2 model. You can quickly start using it in both PyTorch and JAX environments.

✨ Features

• Multimodal Autoregressive Pre - training: AIMv2 is pre - trained with a multimodal autoregressive objective, which is simple and effective for training and scaling.
• High Performance:
  • Outperforms OAI CLIP and SigLIP on the majority of multimodal understanding benchmarks.
  • Outperforms DINOv2 on open - vocabulary object detection and referring expression comprehension.
  • Achieves strong recognition performance, e.g., AIMv2 - 3B reaches 89.5% on ImageNet using a frozen trunk.

📦 Model Information

Model Performance

The following table shows the accuracy of the aimv2 - large - patch14 - 224 model on different classification datasets:

💻 Usage Examples

Basic Usage

PyTorch

import requests
from PIL import Image
from transformers import AutoImageProcessor, AutoModel

url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)

processor = AutoImageProcessor.from_pretrained(
    "apple/aimv2-large-patch14-224",
)
model = AutoModel.from_pretrained(
    "apple/aimv2-large-patch14-224",
    trust_remote_code=True,
)

inputs = processor(images=image, return_tensors="pt")
outputs = model(**inputs)

JAX

import requests
from PIL import Image
from transformers import AutoImageProcessor, FlaxAutoModel

url = "http://images.cocodataset.org/val2017/000000039769.jpg"
image = Image.open(requests.get(url, stream=True).raw)

processor = AutoImageProcessor.from_pretrained(
    "apple/aimv2-large-patch14-224",
)
model = FlaxAutoModel.from_pretrained(
    "apple/aimv2-large-patch14-224",
    trust_remote_code=True,
)

inputs = processor(images=image, return_tensors="jax")
outputs = model(**inputs)

📄 License

This project is licensed under the apple - amlr license.

📚 Documentation

For more details about the AIMv2 model, please refer to the [AIMv2 Paper].

📖 Citation

If you find our work useful, please consider citing us as:

@misc{fini2024multimodalautoregressivepretraininglarge,
  author      = {Fini, Enrico and Shukor, Mustafa and Li, Xiujun and Dufter, Philipp and Klein, Michal and Haldimann, David and Aitharaju, Sai and da Costa, Victor Guilherme Turrisi and Béthune, Louis and Gan, Zhe and Toshev, Alexander T and Eichner, Marcin and Nabi, Moin and Yang, Yinfei and Susskind, Joshua M. and El-Nouby, Alaaeldin},
  url         = {https://arxiv.org/abs/2411.14402},
  eprint      = {2411.14402},
  eprintclass = {cs.CV},
  eprinttype  = {arXiv},
  title       = {Multimodal Autoregressive Pre-training of Large Vision Encoders},
  year        = {2024},
}