Who leads the ARC-AGI-2 benchmark?

Gemini 2.5 Pro currently leads ARC-AGI-2 with a score of 5.00 on Accuracy.

What is the state-of-the-art score on ARC-AGI-2?

The state-of-the-art result on ARC-AGI-2 is 5.00 (Accuracy), achieved by Gemini 2.5 Pro as of 2026.

How many models are tracked on ARC-AGI-2?

Codesota tracks 3 models on ARC-AGI-2.

When was the ARC-AGI-2 leaderboard last updated?

The ARC-AGI-2 leaderboard on Codesota includes results through 2026.

Codesota · Benchmark · ARC-AGI-2Home/Leaderboards/Language & Knowledge/Logical Reasoning/ARC-AGI-2

Unknown

ARC-AGI-2.

Name: ARC-AGI-2 Benchmark Results
Creator: Unknown
Published: 2026-01-01
License: https://creativecommons.org/licenses/by/4.0/

Harder successor to ARC-AGI-1, released 2025. Designed to be more resistant to test-time compute scaling. Scores reported as % on public evaluation set.

Paper ↗Leaderboard ↓

§ 01 · Leaderboard

Results by metric.

Only 3 models on this benchmark

Help build the community leaderboard — submit your model results.

Found a wrong score or missing run?

Use row edits to send a sourced correction into moderation.

Add / edit result ↗Report issue ↗

Accuracy

Accuracy is the reported evaluation metric for ARC-AGI-2. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Accuracyverifiedpapervendorcommunityunverified

Rank	Model	Trust	Score	Year	Links	Fix
01	Gemini 2.5 Pro Public evaluation set. Source: ARC Prize leaderboard (2025).	verified	5.00	2026	Source ↗	Looks wrong?
02	o3 Public evaluation set. Released May 2025. Much harder than v1 — o3 only scores ~4%. Source: ARC Prize blog (2025).	verified	4.00	2026	Source ↗	Looks wrong?
03	o4-mini Public evaluation set. Source: ARC Prize leaderboard (2025).	verified	3.00	2026	Source ↗	Looks wrong?

§ 04 · Submit a result

Add to the leaderboard.

← Back to Logical Reasoning