mbreuss/consistency_trajectory_models_toy_task

Minimal unofficial implementation of Consistency Trajectory models on a 1D toy task.

22
/ 100
Experimental

This project offers a minimal, unofficial implementation of Consistency Trajectory Models on a simple 1D task. It allows researchers to explore how these generative models learn to predict probability flow ODE trajectories directly, differing from traditional diffusion models. The output demonstrates how these models perform at generating data based on learned trajectories, and it's intended for machine learning researchers and academics studying generative AI.

No commits in the last 6 months.

Use this if you are a machine learning researcher or student who wants to understand and experiment with the core concepts of Consistency Trajectory Models on a basic, controlled dataset.

Not ideal if you need a robust, production-ready implementation for high-resolution image generation or real-world applications.

Generative AI Research Machine Learning Theory Diffusion Models AI Model Prototyping Probabilistic Modeling
Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 6 / 25
Maturity 16 / 25
Community 0 / 25

How are scores calculated?

Stars

22

Forks

Language

Python

License

MIT

Last pushed

Mar 11, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/mbreuss/consistency_trajectory_models_toy_task"

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