Pheye-x4-448

miguelcarv/Pheye-x4-448

Pheye-x4-448 is a image-text-to-text model from miguelcarv. Use it for the image-text-to-text task on the model card, and read the license before you ship it in a product. It is set up for transformers.

- These models train a fraction of the number of parameters other models of similar sizes train - Are more efficient in how they incorporate vision into language tasks (dense cross-attention LLaVA style architecture)…

Downloads · 30 days

7% of all-time downloads

All-time downloads

109

Public

Repo size

10 GB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.pt5.3 GB · 100%

At a glance

Task: Image-Text-to-Text
Library: transformers
Access: Public
Created: May 7, 2024
Updated: Jan 2, 2025
SHA: 8764817f

Try a prompt

Model card

From the Hugging Face model README

Φ Pheye - a family of efficient small vision-language models

These models train a fraction of the number of parameters other models of similar sizes train
Are more efficient in how they incorporate vision into language tasks (dense cross-attention > LLaVA style architecture)
Are more efficient in how the process high resolution input images
Use less data to train yet achieve competitive results (if you want to redo this recipe using more data results will be even better)

Benchmarks

Model	Resolution	Trainable Params	Data	VQAv2	Nocaps	TextVQA	TextCaps
MoE-LLaVA-1.6B×4	336	2.9B	5.9M	76.0	-	47.8	-
MoE-LLaVA-2.7B×4	336	5.3B	5.9M	77.1	-	50.2	-
moondream1	384	1.86B	3.9M	74.7	-	35.6
moondream2	384	1.86B	-	77.7	92.5	49.7	120.2
Pheye-x4 🤗	448	295M	2.9M	75.2	110.1	45.9	106.4
Pheye-x4 🤗	672	295M	2.9M	75.5	110.8	49.2	111.9
Pheye-x2 🤗	448	578M	2.9M	76.0	111.8	47.3	108.9
Pheye-x2 🤗	672	578M	2.9M	76.4	110.5	50.5	115.9

Examples*

Image	Example
<img src="https://c5.staticflickr.com/6/5463/17191308944_ae0b20bb7e_o.jpg" width="500"/>	How much do these popcorn packets weigh?<br>These popcorn packets weigh 90g each.<br><br>What brand are these popcorn packets?<br>These popcorn packets are from the brand "Magic Pop".
<img src="https://farm2.staticflickr.com/2708/5836100440_6e1117d36f_o.jpg" width="500"/>	Can I pet that dog?<br>No, you cannot pet the dog in the image. It is a black bear, not a dog.
<img src="images/beekepper.jpeg" width="500"/>	Why is the person wearing protective gear?<br>The person in the image is wearing protective gear, such as a beekeeper's suit and a helmet, to ensure their safety while working with bees. Beekeeping involves handling bees and their hives, which can be potentially dangerous due to the risk of bee stings. The protective gear helps to minimize the risk of bee stings and other bee-related injuries, allowing the beekeeper to carry out their work safely and effectively.

* Generated by Pheye-x2-672

Usage

To generate a sample response from a prompt use generate.py. Use a Python version >= 3.11. Start by cloning the repo and create a virtual environment with the necessary packages:

git clone https://github.com/miguelscarv/pheye.git
cd pheye
python3 -m venv venv
source venv/bin/activate
pip3 install -r requirements.txt

Then run generate.py:

python3 generate.py --image_path images/dog_flower.jpg --prompt "What is the dog holding in it's mouth?" --device cuda

Acknowledgments

This implementation was inspired by OpenFlamingo's repository.

Task: Image-Text-to-Text
Library: transformers
Languages: en
Created: May 7, 2024
Updated: Jan 2, 2025