HomeProjectsArticlesPK BoardAbout
Explore Projects

OpenClaw-RL

Stale
GitHub Python Apache-2.0

Description

OpenClaw-RL: Train any agent simply by talking.

Key Features

  • Fully asynchronous 4-component RL architecture decoupling agent serving, rollout collection, evaluation, and policy training
  • Self-hosted and private by design — entire stack runs on your infrastructure with no third-party API dependency
  • Automatic training from natural conversation feedback with zero manual labeling or dataset curation
  • Three optimization methods: Binary RL (GRPO), On-Policy Distillation (OPD), and Hybrid combining both approaches
  • Supports both personalized agent optimization and scalable RL for terminal, GUI, SWE, and tool-call agents
  • Cloud and local GPU deployment via Tinker integration with one-line launch and LoRA training support

Use Cases

💡 Training personalized AI agents by simply talking to them — collecting conversation feedback as natural training signals
💡 Optimizing terminal, GUI, and coding agents in real-world environments with scalable environment parallelization
💡 Improving agent performance on specific tasks through continuous reinforcement learning during production use
💡 Building self-improving AI systems that learn from user interactions without explicit reward engineering
💡 Research on hybrid RL methods combining scalar rewards with token-level directional signals for robust optimization

Strengths & Limitations

✅ Strengths

  • • High community interest (5.7k stars)
  • • Permissive open-source license (Apache-2.0)

Tags

reinforcement-learning openclaw skill-learning python

Categories

🤖 Agent Framework

Quick Start

Install the OpenClaw-RL extension for your OpenClaw instance, configure your local GPU or Tinker cloud endpoint, and launch with a single command. The framework starts collecting conversation data and optimizing your agent model in the background while you continue using it.
Visit GitHub

Related Projects

ART

10.7k · Python
Active A+

ART (Agent Reinforcement Trainer) trains multi-step agents for real-world tasks using GRPO reinforcement learning, enabling on-the-job training for models like Qwen, Llama, and more.

reinforcement-learningagent-trainingGRPO +3
  • · GRPO-based reinforcement learning for training multi-step agents on real-world tasks
  • · W&B Training serverless RL with 40% lower cost and 28% faster training via shared inference cluster
  • · Pre-built notebooks for email search, 2048, Tic Tac Toe, Codenames, and MCP tool mastery
View full details → View on GitHub

Clawith

4.1k · Python
Active A

Clawith is an open-source project for building and operating AI agent systems.

multi-agentopenclawenterprise +1
  • · Aware autonomous consciousness system: agents have adaptive autonomous perception, decision-making and execution capabilities, dynamically creating and adjusting triggers (cron/once/interval/poll/on_message/webhook — six types) for autonomous task scheduling
  • · Persistent identity and workspaces: each agent has a soul.md (personality), memory.md (long-term memory) and independent filesystem, supports sandboxed code execution, identity persists across sessions
  • · Organization-level collaboration: multi-tenant RBAC access control, channel integration (Slack/Discord/Feishu/Lark), approval workflows, audit logs, agents can send messages and delegate tasks across the organization
View on GitHub

OpenRLHF

10.0k · Python
Active A

OpenRLHF is a high-performance agentic RL framework based on Ray and vLLM, offering PPO, DAPO, and REINFORCE++ algorithms for large-scale training of agents and vision-language models.

reinforcement-learningagent-trainingPPO +3
  • · First production-ready open-source RLHF framework built on Ray + vLLM distributed architecture
  • · Supports PPO, DAPO, REINFORCE++, GRPO, and RLOO reinforcement learning algorithms
  • · Hybrid engine scheduling where all models share GPUs to minimize idle time
View full details → View on GitHub