on-policy
★ 2.0k
vs
claude-code-source-all-in-one
★ 80
on-policy vs claude-code-source-all-in-one
on-policy: This repository implements MAPPO, a multi-agent variant of PPO, widely used in cooperative multi-agent games and research. It provides robust implementations for various multi-agent environments like StarCraft II, Hanabi, and Google Research Football, along with detailed training scripts and hyperparameter guidance.; claude-code-source-all-in-one: This repository extracts the source code of Anthropic's Claude Code CLI for educational study. It includes 18 deep-dive articles analyzing the architecture, covering core agent loop, tool orchestration, context compression, and more. The source can be run locally for learning purposes.
01
TL;DR
Choose on-policy if…Research and experimentation in cooperative multi-agent reinforcement learning
Choose claude-code-source-all-in-one if…Studying production-level AI agent architecture and design decisions
02
Side-by-Side Comparison
03
Features
on-policy01Implementation of MAPPO (Multi-Agent PPO)
02Support for diverse multi-agent environments (e.g., StarCraft II, Hanabi)
03Ready-to-use training scripts for various scenarios
04Detailed hyperparameter guidance and updated results
05Default support for shared policy among agents
claude-code-source-all-in-one01Extracted production-grade AI agent source code
02Loop-over-graph architecture with while(true) agent loop
03Recursive sub-agent orchestration with inherited features
04Four-layer context compression for long sessions
05Immutable messages to optimize prompt caching costs
04
Use Cases
on-policy↳Research and experimentation in cooperative multi-agent reinforcement learning
↳Benchmarking and evaluating PPO's effectiveness in MARL scenarios
↳Training AI agents for popular multi-agent games like StarCraft II and Hanabi
claude-code-source-all-in-one↳Studying production-level AI agent architecture and design decisions
↳Learning advanced engineering patterns for tool orchestration and context management
↳Experimenting with Claude Code internals in a controlled environment
05
Best For
on-policyTrendingReinforcement LearningMulti-Agent AI
claude-code-source-all-in-oneTrendingAPI IntegrationDev Tooling
FAQ
FAQ
What is the difference between on-policy and claude-code-source-all-in-one?
Both on-policy and claude-code-source-all-in-one are in the LLM Infra category. on-policy has 2.0k stars, while claude-code-source-all-in-one has 80 stars.
Which is better, on-policy or claude-code-source-all-in-one?
The best choice depends on your use case. Choose on-policy if Research and experimentation in cooperative multi-agent reinforcement learning, and claude-code-source-all-in-one if Studying production-level AI agent architecture and design decisions.
Is on-policy free or open source?
Yes, on-policy is open source on GitHub (MIT).
Is claude-code-source-all-in-one free or open source?
Yes, claude-code-source-all-in-one is open source on GitHub.
→