Slownite/Trident-WM

A high-performance World Model for robotics using PyTorch Lightning. It implements a V-M-D architecture (Vision, Memory, Decoder) for latent-space imagination and video "dreaming" on LeRobot datasets.

19
/ 100
Experimental
No Package No Dependents
Maintenance 10 / 25
Adoption 0 / 25
Maturity 9 / 25
Community 0 / 25

How are scores calculated?

Stars

—

Forks

—

Language

Python

License

MIT

Category

world-models-frameworks

Last pushed

Feb 15, 2026

Commits (30d)

0

GitHub
World Models Frameworks · 25 frameworks

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/Slownite/Trident-WM"

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

Higher-rated alternatives

danijar/dreamerv3

Mastering Diverse Domains through World Models

66

NM512/dreamerv3-torch

Implementation of Dreamer v3 in pytorch.

60

galilai-group/stable-worldmodel

Reliable, minimal and scalable library for evaluating and conducting world model research

59

danijar/dreamerv2

Mastering Atari with Discrete World Models

58

danijar/dreamer

Dream to Control: Learning Behaviors by Latent Imagination

56

Explore ML Frameworks

All categories Trending ML Framework directory Insights