#116116th of 168 in ML frameworks and libraries
open_spiel
Game environments and algorithms for reinforcement learning and search research
- Stars
- 5.6k
- License
- Apache-2.0
- Last commit
- Nov 2026
- Last release
- Aug 2026
- Language
- Python
5,128 on 21 Apr 20265,588 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
OpenSpiel is a DeepMind library of game environments and RL/search algorithms covering n-player, zero-sum, cooperative and general-sum games, with perfect and imperfect information, turn-based and simultaneous moves. The core API and games are written in C++ and exposed to Python, and it includes tools for analyzing learning dynamics and evaluation metrics. Pretrained baseline checkpoints and game datasets are published on a public storage bucket.
Who it is for: RL and game-theory researchers studying multi-agent learning and search
Strengths
- Wide game coverage: extensive-form, simultaneous-move, imperfect-information, grid worlds, social dilemmas
- C++ core with Python bindings; algorithms available in both languages
- Colab notebooks and video tutorials cover CFR and REINFORCE on Kuhn and Leduc poker
- Semantic versioning; pretrained checkpoints and datasets published
Weaknesses
- Research library for games, not a general LLM or agent toolkit
- Separate install paths for Linux/macOS and Windows; core is C++
- README does not list games or algorithms; readers must consult the docs
- No turnkey app; using the environments requires writing code
What it needs
- Docker
Quick answers
Is open_spiel open source?
Yes. Its code is on GitHub under the Apache-2.0 license.
Can you run open_spiel with Docker?
Yes. Its repository has a Docker setup: a Dockerfile, a Compose file or a published image.
Also in ML frameworks and libraries
See all 168| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd | 93 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and training scripts | 93 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 92 out of 100 |