Tiers:
Easy/Med/Hard
Step into the future of Artificial Intelligence with DIAMBRA.
Train your agents on cutting-edge Reinforcement Learning environments and make them compete in live video games tournaments.
Join Our Global Leaderboard
Compete against coders from all around the world
Unlock exclusive achievements and collect them all!
Glorious Agent
"Complete the game suffering a maximum amount of damage equal or lower than 20% of your total cumulative health available"
Games-Based RL Environments
Install DIAMBRA Arena
Diambra Arena transforms popular video games into RL environments! It’s built on the OpenAI Gym Python interface, and we’ve made it very easy to use.
Getting started in three easy moves:
- 1) Install DIAMBRA Arena directly through Python PIP as explained in the Documentation (Linux, Win and MacOS supported)
- 2) Download ready-to-use Examples from DIAMBRA GitHub Repo
- 3) Join DIAMBRA Discord Server to become part of the Community and to receive support
Features
Standard RL
Competitive Multi-Agent
Competitive Human-Agent
SelfPlay
Imitation Learning
Human-in-the-loop Training
Supported Games
Diambra already supports six classic games out of the box, and we are always adding more!
Dead or Alive ++
Street Fighter III
Tekken Tag Tournament
Ultimate Mortal Kombat 3
Samurai Showdown 5 Special
The King of Fighters '98
About Us
DIAMBRA combines high quality Reinforcement Learning environments with a proprietary tournaments platform where coders from all around the world make their AI agents compete in video games.
The history behind DIAMBRA name
Naples, 1552. Two women, Isabella of Carazzi and DIAMBRA of Pottinella, dispute in a duel for the love of a man called Fabio Zeresola. This legendary episode, subject matter of the painting "Women Gladiators", by Jusepe de Ribera, is also seen as an allegory of the fight between Vice and Virtue.