Skip to content

The Aggregate

Unified LLM rankings, daily

Models What's New Trends Benchmarks Skill Maps
How It Works Guesswork Metrics
The Aggregate
Loading data...

Google Gemini 3 Deep Think - ARC-AGI-2 — leaderboard

Metric: Score (%). Source: storage.googleapis.com. 4 models tracked.

Top models

#ModelScore
1Gemini 3 Deep Think84.6
2Claude Opus 4.6 (Thinking)68.8
3GPT-5.2 (xHigh)52.9
4Gemini 3 Pro (Preview)31.1

Interactive version: theaggregate.ai/benchmark?slug=google-gemini-3-deep-think-arc-agi-2 · How the rankings work · Data refreshed daily, snapshot 2026-07-22.

Built by Mikhail Doroshenko — AI researcher, co-author of Humanity’s Last Exam. Independent project; no affiliation with any AI lab.