danijar/dreamerv3
dreamerv3
Mastering Diverse Domains through World Models
Builder
danijar
danijar • individual
Stars
3,031
Using upstream star count
Forks
500
Using upstream fork count
Open Issues
0
Activity Score
0/100
0 commits in 30d
Created
Jan 14, 2023
Project creation date
README Summary
DreamerV3 is a reinforcement learning algorithm that learns world models to master diverse domains through model-based planning. It achieves state-of-the-art performance across various environments including Atari, DMC, BSuite, and Crafter by learning accurate world models and using them for policy optimization. The implementation provides a scalable and general-purpose approach to reinforcement learning that works across different types of tasks and observation modalities.
AI Dev Skills
Unmapped
Tags
Taxonomy
Modalities
Recent Activity
Updated 6 months ago
7 Days
0
30 Days
0
90 Days
0
Quality
research- Quality
- high
- Maturity
- research
Categories
PM Skills
Languages
Timeline
- Project created
- Jan 14, 2023
- Forked
- Mar 31, 2026
- Your last push
- 6 months ago
- Upstream last push
- 6 months ago
- Tracked since
- Sep 23, 2025
Similar Repos
pgvector cosine similarity · $0
Loading…