RFL
Robot Football League (RFL) is an open, daily-streamed competitive league of simulated two-a-side Unitree G1 humanoid robots. It provides an open-source engine, rules, reference clubs and public match telemetry so developers, researchers and teams can run matches locally and qualify to have their code run on real robots.
RFL is gaming software teams evaluate for software & gaming. Use this page to review pricing, integration signals, and the best alternatives before you commit.
Profile facts come from the vendor source. AiMatch labels unknown pricing or API details instead of estimating them.
Review official source →Quick Overview
Best for: Software & Gaming
What it does
Gaming software for decision-makers comparing workflow fit and alternatives.
Best fit
Software & Gaming
Pricing snapshot
Free from Free
Next step
Compare RFL with similar tools before you shortlist it.
Compare this tool before you shortlist it
Review alternatives, pricing posture, and workflow fit side by side.
RFL
RFL (Robot Football League) runs a daily, open competition of simulated two-a-side Unitree G1 humanoid robots using MuJoCo physics and a shared player stack shape similar to RoboCup humanoid teams. The project publishes its engine, rules, reference clubs, match logs, telemetry and every club's commit history so anyone can reproduce matches locally, enter teams, and audit play. The simulation functions as a qualifying round: the stated goal is to progress to live matches where winning code is run on real Unitree G1 hardware on a real pitch.
The Robot Football League: 2v2 simulated Unitree G1 humanoids playing competitive football, three matches a day, streamed on Twitch and in spatial 4DGSx. The engine, the rules and every club are open source — anyone can enter a team.
Own this listing?
Claim this page for a one-time $29 to add pricing, features, screenshots, verified owner details, and a clearly labeled 30-day category position after the profile is live.
Claim this listing for $29Key Features
Open-source engine & rules
The league engine and rules are public and MIT-licensed, with a working reference club available on GitHub.
Simulated Unitree G1 humanoids in MuJoCo
Two-a-side Unitree G1 humanoids simulated in real physics (MuJoCo), walking on a pretrained policy and using the same stack shape as RoboCup humanoid teams.
Real-robot constraints enforced
Players see only what a real G1 could see (onboard camera and team radio), use a 2-second decision cycle, and motion goes through the interfaces a real G1 exposes.
Full transparency and auditability
Every club's code, match logs, telemetry, radio communications and commit histories are published and visible.
Daily streamed matches and 3D replays
Matches are streamed daily and replays can be watched in 3D (4DGSx spatial streaming and standard streams like Twitch/YouTube).
Reference club & quick start
A reference club and demo brains are provided; example commands show how to run a match locally in minutes.
Pricing
Yes — the engine, rules and reference club are open-source (MIT) and matches are free to watch.
Open / Free
Free- Engine and rules MIT-licensed
- Free-to-watch daily streams
- Reference club and example demo brains included
Use Cases
Research and development in humanoid robotics
Researchers can use the MuJoCo-simulated Unitree G1 environment and published telemetry to develop perception, control and multi-agent strategies.
University and robotics club competitions
Universities and robotics groups can fork the reference club, modify behaviours and enter teams in an open league designed to be reproducible on real robots.
AI agent development and benchmarking
Teams can test frontier AI models or custom controllers in daily matches and publicly compare performance using published logs and commit histories.
Spectatorship and education
Matches are free to watch and streamed daily, with 3D replays that support learning about robot behaviour and match dynamics.
Integrations
MuJoCo physics
Simulation of Unitree G1 humanoids uses MuJoCo physics for real-physics behaviour.
Unitree G1 robot model
The robot model (Unitree G1) is used as the simulation target; model copyright noted on the site.
GitHub (engine & code hosting)
Engine, reference club and full commit history are hosted on GitHub for cloning, development and auditing.
Streaming platforms (4DGSx, Twitch, YouTube)
Matches are streamed daily with support for 4DGSx spatial 3D viewing and standard streaming platforms.
Benefits
Limitations
Frequently Asked Questions
No verified FAQs are available.
Getting Started
- 1 Clone the engine repository: git clone https://github.com/robot-football-league/rfl-engine
- 2 Install the engine: cd rfl-engine && pip install -e .
- 3 Run a demo match: python -m gauntlet rfl examples/mock_united \ examples/mock_united --video match.mp4
- 4 Fork the reference club, modify behaviour, and enter your team following the league rules
Support
docs
Engine and examples are available on GitHub: https://github.com/robot-football-league/rfl-engine
contact
General contact page on the site: https://rfl.football/contact
streams & community
Watch matches and follow broadcasts via the site links to Twitch, YouTube and 4DGSx for community interaction
API
https://github.com/robot-football-league/rfl-engine
Compare RFL with similar tools
See how it stacks up against alternatives
Related Tools
View all 13 →
Scrabblecam
Scrabblecam is a photo-based Scrabble solver that analyzes pictures of Scrabble boards and suggests best moves across multiple dictionary variants.
cascadeur
Cascadeur is a standalone 3D keyframe animation application with AI-assisted and physics tools for creating, cleaning up, retargeting and editing character animation. It supports .FBX, .DAE and .USD workflows and provides rigging, AutoPosing, AutoPhysics and mocap-cleaning features for hobbyists and professionals.
rebellis-ai
Rebellis AI is an AI-powered text-to-3D animation platform that generates realistic character motions from text prompts, producing up to three high-quality 3D animations in about 30 seconds and offering plugins and an API for integration into game and animation pipelines.