ByteVisionLab/TokenFlow

[CVPR 2025] 🔥 Official impl. of "TokenFlow: Unified Image Tokenizer for Multimodal Understanding and Generation".

36
/ 100
Emerging

This project offers a powerful tool for AI developers working with both image understanding and image generation tasks. It takes raw images or text prompts as input and outputs highly optimized image tokens that can be used to improve the performance of large multimodal models (LMMs) for tasks like visual question answering or text-to-image creation. It's ideal for machine learning engineers and researchers focused on building or enhancing AI models that interpret and create visual content.

449 stars. No commits in the last 6 months.

Use this if you are developing AI models that need to both understand and generate images, and you want a unified approach to process visual data for these tasks.

Not ideal if you are looking for an end-user application or a tool for image editing, as this is a foundational component for AI model development.

AI model development computer vision engineering multimodal AI generative AI image recognition
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 8 / 25

How are scores calculated?

Stars

449

Forks

9

Language

Python

License

Apache-2.0

Last pushed

Aug 08, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/ByteVisionLab/TokenFlow"

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