ARC Prize 2026: ARC-AGI-3 Milestone Prize #1

ARC Prize 2026: ARC-AGI-3 Milestone Prize #1

The top three submissions from ARC-AGI-3 Milestone 1

ARC-AGI-3 is our first Interactive Reasoning Benchmark. It is a set of novel, video-game-like environments where agents must perceive, explore, plan, and act across long horizons. Due to a hidden hold-out set, these games can't be memorized. They reward on-the-fly learning, efficient exploration, and the ability to set your own goals when the objective isn't clear. Solving them is easy (and even fun) for humans, but remains hard for today's AI.

New to ARC Prize 2026, we've introduced two milestone prizes that reward the top mid-competition open-source solutions to ARC-AGI-3. They exist to inspire people to try new ideas and create incentives to push those ideas further.

We're excited to award our first $37.5K ARC-AGI-3 milestone prize, which ran through June 30th. Below are the top three submissions from our ARC-AGI-3 Kaggle competition.

1st Place - Tufa Labs, "The Duck"

A small open-source LLM that plays ARC-AGI-3 by writing and running Python in a live REPL (similar to the Duke Harness), treating each game like an interactive coding problem.

Notebook · Write-up · Video · Tufa Labs

{/ Crop-ready: to crop, set a on this wrapper and adjust on the img below /}

This submission converts the game state into Python variables and interacts with it through a REPL. The process includes reasoning, calling helper functions, running code, taking an action, seeing the results, repeating. It perceives the board through a rendered image, the raw ASCII grid, and a segmentation tool for zooming into regions. It chooses whichever representation fits the moment.

To keep playing indefinitely without exhausting the context window, The Duck continuously pops the oldest messages ("infinite play via eviction"), keeping only the system prompt and recent history. The team's stated philosophy is to keep the harness lightweight and generic and let…