---
license: apache-2.0
tags:
- vision
---

# SigLIP 2 So400m

[SigLIP 2](https://huggingface.co/collections/google/siglip2-67b5dcef38c175486e240107)
extends the pretraining objective of
[SigLIP](https://huggingface.co/collections/google/siglip-659d5e62f0ae1a57ae0e83ba)
with prior, independently developed techniques into a unified recipe, for improved semantic
understanding, localization, and dense features.

## Intended uses

You can use the raw model for tasks like zero-shot image classification and
image-text retrieval, or as a vision encoder for VLMs (and other vision tasks).

Here is how to use this model to perform zero-shot image classification:

```python
from transformers import pipeline

# load pipeline
ckpt = "google/siglip2-so400m-patch16-512"
image_classifier = pipeline(model=ckpt, task="zero-shot-image-classification")

# load image and candidate labels
url = "http://images.cocodataset.org/val2017/000000039769.jpg"
candidate_labels = ["2 cats", "a plane", "a remote"]

# run inference
outputs = image_classifier(image, candidate_labels)
print(outputs)
```

You can encode an image using the Vision Tower like so:

```python
import torch
from transformers import AutoModel, AutoProcessor
from transformers.image_utils import load_image

# load the model and processor
ckpt = "google/siglip2-so400m-patch16-512"
model = AutoModel.from_pretrained(ckpt, device_map="auto").eval()
processor = AutoProcessor.from_pretrained(ckpt)

# load the image
image = load_image("https://huggingface.co/datasets/merve/coco/resolve/main/val2017/000000000285.jpg")
inputs = processor(images=[image], return_tensors="pt").to(model.device)

# run infernece
with torch.no_grad():
    image_embeddings = model.get_image_features(**inputs)    

print(image_embeddings.shape)
```

For more code examples, we refer to the [siglip documentation](https://huggingface.co/transformers/main/model_doc/siglip.html#).

## Training procedure

SigLIP 2 adds some clever training objectives on top of SigLIP:

1. Decoder loss
2. Global-local and masked prediction loss
3. Aspect ratio and resolution adaptibility 

### Training data

SigLIP 2 is pre-trained on the WebLI dataset [(Chen et al., 2023)](https://arxiv.org/abs/2209.06794).

### Compute

The model was trained on up to 2048 TPU-v5e chips.

## Evaluation results

Evaluation of SigLIP 2 is shown below (taken from the paper).

[Evaluation Table](TODO)

### BibTeX entry and citation info

```bibtex
TODO
```