AgentIndex icon
AgentIndex
verl-agent logo
verl-agent
★ 2.4k
vs
on-policy logo
on-policy
★ 2.1k

verl-agent vs on-policy

verl-agent: `verl-agent` extends veRL to train LLM agents using reinforcement learning, featuring a novel step-independent multi-turn rollout mechanism. This design ensures high scalability for long-horizon tasks by allowing customizable per-step input structures and memory management.; 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.

01

TL;DR

verl-agent logo
verl-agentLLM Infra · ★ 2.4k

Training large language model agents for complex multi-turn, long-horizon tasks.

Read verl-agent full review →
on-policy logo
on-policyLLM Infra · ★ 2.1k

Research and experimentation in cooperative multi-agent reinforcement learning

Read on-policy full review →
02

Side-by-Side Comparison

Field
verl-agent logoverl-agent
on-policy logoon-policy
Category
LLM Infra
LLM Infra
Stars
★ 2.4k
★ 2.1k
License
Apache-2.0
MIT
Install Command
pip install -e .
pip install -e .
Updated
4mo ago
2y ago
Open Source
Yes
Yes
Website
↗ Visit
↗ Visit
GitHub
↗ GitHub
↗ GitHub
Tags
LLM Agents, Reinforcement Learning, Deep Learning
Multi-Agent Reinforcement Learning, PPO, MAPPO
03

Features

verl-agent logoverl-agent
01Multi-Turn Agent-Environment Interaction
02Fully Customizable Memory Module & Per-Step Input Structure
03Scalable for Very Long-Horizon Optimization
04Parallelized Gym-Style Environments and Group Environments
05Diverse Reinforcement Learning Algorithms
on-policy logoon-policy
01Implementation 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
04

Use Cases

verl-agent logoverl-agent
↳Training large language model agents for complex multi-turn, long-horizon tasks.
↳Developing reasoning agents for both visual and text-based environments.
↳Solving digital interface control, embodied AI, and search-related challenges.
on-policy logoon-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
FAQ

FAQ

What is the difference between verl-agent and on-policy?
Both verl-agent and on-policy are in the LLM Infra category. verl-agent has 2.4k stars, while on-policy has 2.1k stars.
Which is better, verl-agent or on-policy?
The best choice depends on your use case. Choose verl-agent if Training large language model agents for complex multi-turn, long-horizon tasks., and on-policy if Research and experimentation in cooperative multi-agent reinforcement learning.
Is verl-agent free or open source?
Yes, verl-agent is open source on GitHub (Apache-2.0).
Is on-policy free or open source?
Yes, on-policy is open source on GitHub (MIT).
Cloud Infrastructure
Ad

Deploy on DigitalOcean

Host your MCP servers, Docker containers, and AI agents with $200 free credit over 60 days.

Claim $200 Free Credit
→

Related

Alternatives to verl-agent →Alternatives to on-policy →verl-agent details →on-policy details →