从 65,729 个高星开源项目里，
找到你要的下一个。

RL4LMs 是用 Python 编写的学习资料，用于系统学习与查阅。

A modular RL library to fine-tune language models to human preferences

gym-anytrading 是用 Python 编写的学习资料，用于系统学习与查阅，主打易用、功能完整、可扩展。

The most simple, flexible, and comprehensive OpenAI Gym trading environment (Approved by OpenAI Gym)

PPO-PyTorch 是用 Python 编写的学习资料，用于系统学习与查阅，主打轻量。

Minimal implementation of clipped objective Proximal Policy Optimization (PPO) in PyTorch

MimicKit 是用 Python 编写的学习资料，用于系统学习与查阅，主打轻量。

A lightweight suite of motion imitation methods for training controllers.

verl-agent 是用 Python 编写的 AI 工具，用于大模型与智能体应用开发。

verl-agent is an extension of veRL, designed for training LLM/VLM agents via RL. verl-agent is also the official code for paper "Group-in-Group Poli…

ICCV2019-LearningToPaint 是用 Python 编写的学习资料，用于系统学习与查阅。

ICCV2019 - Learning to Paint With Model-based Deep Reinforcement Learning

drl-zh 是用 Jupyter Notebook 编写的学习资料，用于系统学习与查阅。

Deep Reinforcement Learning: Zero to Hero!

DeepRL-Agents 是用 Jupyter Notebook 编写的学习资料，用于系统学习与查阅，主打实现清晰。

A set of Deep Reinforcement Learning Agents implemented in Tensorflow.

chess-alpha-zero 是用 Jupyter Notebook 编写的学习资料，用于系统学习与查阅。

Chess reinforcement learning by AlphaGo Zero methods.

gym-pybullet-drones 是用 Python 编写的学习资料，用于系统学习与查阅。

PyBullet Gymnasium environments for single and multi-agent reinforcement learning of quadcopter control

Nemotron 是用 Jupyter Notebook 编写的学习资料，用于系统学习与查阅。

Developer Asset Hub for NVIDIA Nemotron — A one-stop resource for training recipes, usage cookbooks, datasets, and full end-to-end reference example…

ASAP 是用 Python 编写的学习资料，用于系统学习与查阅。

[RSS 2025] "ASAP: Aligning Simulation and Real-World Physics for Learning Agile Humanoid Whole-Body Skills"

diamond 是用 Python 编写的 AI 工具，用于人工智能方向的建模与推理。

DIAMOND (DIffusion As a Model Of eNvironment Dreams) is a reinforcement learning agent trained in a diffusion world model. NeurIPS 2024 Spotlight.

ViZDoom 是用 C++ 编写的学习资料，用于系统学习与查阅。

Reinforcement Learning environments based on the 1993 game Doom :godmode:

ELF 是用 C++ 编写的 AI 工具，用于人工智能方向的建模与推理，主打轻量、可扩展。

An End-To-End, Lightweight and Flexible Platform for Game Research

AgentFlow 是用 Python 编写的 AI 工具，用于大模型与智能体应用开发。

AgentFlow: In-the-Flow Agentic System Optimization

rainbow-is-all-you-need 是用 Python 编写的学习资料，用于系统学习与查阅。

Rainbow is all you need! A step-by-step tutorial from DQN to Rainbow

PhySO 是用 Python 编写的学习资料，用于系统学习与查阅。

Physical Symbolic Optimization

all-rl-algorithms 是用 Jupyter Notebook 编写的 AI 工具，用于大模型与智能体应用开发。

Implementation of all RL algorithms in a simpler way

Gymnasium-Robotics 是用 Python 编写的学习资料，用于系统学习与查阅。

A collection of robotics simulation environments for reinforcement learning

TradingGym 是用 Python 编写的交易类项目，用于商城、支付与交易场景，主打易用。

Trading and Backtesting environment for training reinforcement learning agent or simple rule base algo.

RoboVerse 是用 Python 编写的学习资料，用于系统学习与查阅，主打可伸缩。

RoboVerse: Towards a Unified Platform, Dataset and Benchmark for Scalable and Generalizable Robot Learning

snake 是用 Python 编写的学习资料，用于系统学习与查阅。

Playing the game of snake with AI.

Minari 是用 Python 编写的学习资料，用于系统学习与查阅。

A standard format for offline reinforcement learning datasets, with popular reference datasets and related utilities