Skip to content

Latest commit

 

History

69 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 

Repository files navigation

ARC-AGI-3 General Agents

This repository is the project page for our experiments with coding agents designed to be game-general within ARC-AGI-3. We develop and evaluate agent architectures, reasoning strategies, and tools for solving unfamiliar interactive games.

General-agent setting

The agents contain no game-specific code, prompts, hand-coded heuristics, hidden solutions, or other game-specific information. The same agent code and prompts are used across games.

Each game is evaluated as a single recorded playthrough from a fresh workspace. The agent has no access to files, logs, or conversation state from previous playthroughs. It cannot restart the game or return to completed levels.

Public-game saturation

With GPT-5.6-sol, ewma_sv_v1.6 fully solved all 25 public games at both tested reasoning efforts, achieved about 99% mean per-game Relative Human Action Efficiency (RHAE), and used fewer than half as many environment actions as the human baseline.

Because GPT-5.6-sol postdates the public games and held-out performance has not been tested, we interpret this as saturation of the public set—not evidence that ARC-AGI-3 has been solved generally.

Recommended agents

For new experiments, consider:

  • ewma_sv_v1.6: maintains and simplifies a fixed-interface executable world model and verifies it through exact replay of recorded observations. It gives the strongest public-game results but is relatively expensive to run.
  • twma_v1.6: maintains a textual world model without requiring a persistent executable simulator or replay-verification stack. It is much simpler and less expensive while still achieving strong results.

Papers and supporting materials

Citation

If you use this repository or the released results, please cite the relevant paper.

License

Released under the MIT License.

About

No description, website, or topics provided.

Resources

Stars

46 stars

Watchers

2 watching

Forks

Releases

Packages

Contributors

Languages