Agora-Lab-AI/OmniByteGPT

An implementation of an all-new foundation model architecture that trains on byte sequences from multiple modalities to handle omni-modal generation of text, video, images and more.

31
/ 100
Emerging

This project helps AI researchers and advanced practitioners create and manipulate diverse digital content using a single model. It takes in raw byte sequences, which can represent anything from text to video, and generates new byte sequences that can be interpreted as various content types. It's designed for those who need a unified approach to generate and transform text, images, audio, and other data without relying on separate, specialized models for each.

Use this if you are developing new foundation models and need a truly universal architecture for generating content across multiple modalities, from raw byte data.

Not ideal if you are looking for an off-the-shelf solution for single-modality content creation or if you are not comfortable working with byte-level data.

AI-research multi-modal-generation foundation-models universal-data-processing generative-AI
No Package No Dependents
Maintenance 10 / 25
Adoption 5 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

9

Forks

Language

Python

License

MIT

Last pushed

Mar 09, 2026

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/Agora-Lab-AI/OmniByteGPT"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.