Builder
OpenAI
openai • ai-lab
Stars
3,580
Using upstream star count
Forks
533
Using upstream fork count
Open Issues
0
Activity Score
0/100
0 commits in 30d
Created
Feb 7, 2018
Project creation date
README Summary
Retro is an OpenAI Gym wrapper for classic video games, enabling reinforcement learning research on retro gaming environments. It provides a standardized interface for training AI agents on games like Sonic the Hedgehog, Street Fighter, and other classic titles. The library includes pre-configured game scenarios and reward functions for common RL benchmarks.
AI Dev Skills
Unmapped
Reinforcement LearningGame Environment SimulationMulti-Agent SystemsDeep Q-NetworksPolicy Gradient MethodsOpenAI Gym IntegrationGame AIEmulation Interfaces
Tags
Reinforcement LearningGame Environment SimulationMulti-Agent SystemsDeep Q-NetworksPolicy Gradient MethodsOpenAI Gym IntegrationGame AIEmulation InterfacesGame AI BenchmarkingResearch EnvironmentSelf-hostedClassic Game AutomationEntertainmentMulti-Game Agent DevelopmentImageGamingAgentic AIReinforcement Learning ResearchAI Game Agent TrainingResearchAI GamingVideoC
Taxonomy
Deployment Context
Industries
Skill Areas
Recent Activity
Updated 2 years ago
7 Days
0
30 Days
0
90 Days
0
Quality
research- Quality
- medium
- Maturity
- research
Categories
Foundation ModelsPrimaryAI AgentsModel TrainingEvals & BenchmarkingDev Tools & AutomationLearning ResourcesCoding & Dev ToolsSearch & KnowledgeOther AI / MLRobotics
PM Skills
Developer Platform
Languages
C100.0%
Timeline
- Project created
- Feb 7, 2018
- Forked
- Mar 14, 2026
- Your last push
- 2 years ago
- Upstream last push
- 2 years ago
- Tracked since
- Feb 22, 2024
Similar Repos
pgvector cosine similarity · $0
Loading…