lucidrains/dreamer4
Implementation of Danijar's latest iteration for his Dreamer line of work
This project provides an implementation of Dreamer 4, a reinforcement learning algorithm that helps train AI agents more efficiently. It takes observed video sequences, rewards, and agent actions as input to build a "world model." This model can then generate imagined scenarios, allowing the AI agent to learn from simulated experiences and improve its decision-making. AI/ML researchers and practitioners focused on reinforcement learning and embodied AI would find this useful for developing intelligent agents.
165 stars. Available on PyPI.
Use this if you are an AI researcher or practitioner looking to train agents effectively by simulating future scenarios and learning from generated 'dreams' within a world model.
Not ideal if you are looking for a plug-and-play solution for general video generation or traditional machine learning tasks outside of reinforcement learning research.
Stars
165
Forks
12
Language
Python
License
MIT
Category
Last pushed
Mar 27, 2026
Commits (30d)
0
Dependencies
16
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/transformers/lucidrains/dreamer4"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related models
lucidrains/x-transformers
A concise but complete full-attention transformer with a set of promising experimental features...
kanishkamisra/minicons
Utility for behavioral and representational analyses of Language Models
lucidrains/simple-hierarchical-transformer
Experiments around a simple idea for inducing multiple hierarchical predictive model within a GPT
Nicolepcx/Transformers-in-Action
This is the corresponding code for the book Transformers in Action
kyegomez/zeta
Build high-performance AI models with modular building blocks