---
license: apache-2.0
pipeline_tag: feature-extraction
tags:
- clip
- vision
datasets:
- Ziyang/yfcc15m
- conceptual_captions
---
<h1 align="center">UForm</h1>
<h3 align="center">
Pocket-Sized Multimodal AI<br/>
For Content Understanding and Generation<br/>
In Python, JavaScript, and Swift<br/>
</h3>

---

The `uform3-image-text-english-large` UForm model is a tiny vision and English language encoder, mapping them into a shared vector space.
This model produces up to __64-, 256-, 512-, and 768-dimensional embeddings__ and is made of:

* Text encoder: 12-layer BERT for up to 64 input tokens.
* Visual encoder: ViT-L/14 for images of 224 x 224 resolution.

Unlike most CLIP-like multomodal models, this model shares 6 layers between the text and visual encoder to allow for more data- and parameter-efficient training.
Also unlike most models, UForm provides checkpoints compatible with PyTorch, ONNX, and CoreML, covering the absolute majority of AI-capable devices, with pre-quantized weights and inference code.
If you need a larger, more accurate, or multilingual model, check our [HuggingFace Hub](https://huggingface.co/unum-cloud/).
For more details on running the model, check out the [UForm GitHub repository](https://github.com/unum-cloud/uform/).

## Evaluation

For zero-shot ImageNet classification the model achieves Top-1 accuracy of 51.8% and Top-5 of 75.6%.
On text-to-image retrieval it reaches 92% Recall@10 for Flickr:

| Dataset |  Recall@1 |  Recall@5 | Recall@10 |
| :-------- | ------: | --------: | --------: |
| Zero-Shot Flickr    | 0.693 | 0.875 | 0.923 |
| Zero-Shot MS-COCO   | 0.382 | 0.617 | 0.728 |

## Installation

```bash
pip install "uform[torch,onnx]"
```

## Usage

To load the model:

```python
from uform import get_model, Modality

import requests
from io import BytesIO
from PIL import Image

model_name = 'unum-cloud/uform3-image-text-english-large'
modalities = [Modality.TEXT_ENCODER, Modality.IMAGE_ENCODER]
processors, models = get_model(model_name, modalities=modalities)

model_text = models[Modality.TEXT_ENCODER]
model_image = models[Modality.IMAGE_ENCODER]
processor_text = processors[Modality.TEXT_ENCODER]
processor_image = processors[Modality.IMAGE_ENCODER]
```

To encode the content:

```python
text = 'a cityscape bathed in the warm glow of the sun, with varied architecture and a towering, snow-capped mountain rising majestically in the background'
image_url = 'https://media-cdn.tripadvisor.com/media/photo-s/1b/28/6b/53/lovely-armenia.jpg'
image_url = Image.open(BytesIO(requests.get(image_url).content))

image_data = processor_image(image)
text_data = processor_text(text)
image_features, image_embedding = model_image.encode(image_data, return_features=True)
text_features, text_embedding = model_text.encode(text_data, return_features=True)
```