#self-play
GitHub repositories that have self-applied the topic "self-play" — a creator-tagged metadata that surfaces how AI projects describe themselves.
REPOS Repos for #self-play (top 4 by stars)
Modern JAX/Flax recreation & expansion of OpenAI's Emergent Tool Use (hide-and-seek): MAPPO + CTDE, entity Transformer + GRU memory, ELO self-play, and a clean Three.js 3D viewer.
GeFAA/kivski-tactical-ai-simulatorTop-down 2D 5v5 bomb-defuse multi-agent RL simulator with live match viewer. Recurrent MAPPO + TarMAC emergent communication, no scripted strategies.
study8677/openrsiOpen-source Recursive Self-Improvement (RSI) engine for AI Agents & LLMs: Harness-RSI (self-rewriting agent code in sandboxes) + Model-RSI (verifiable self-play & RLVR post-training). 开源递归自我改进引擎:沙盒自改代码 + 模型自进化。
daguix/tarot-botFour-player French Tarot with a Rust rules engine, self-play reinforcement learning agents, and a React interface
RELATED Other topics · full topics ranking →
#claude-code
3,808#ai-agents
3,273#llm
2,099#mcp
1,872#claude
1,861#python
1,690#developer-tools
1,537#codex
1,382#ai
1,186Aggregated by case-insensitive match against topics_json of each repo's latest content snapshot. methodology