OpenCodePapers

abstract-reasoning-on-arc-agi-2

Visual ReasoningAbstract Reasoning
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
PaperCodeAccuracyModelNameReleaseDate
GPT-6 Astra: A new generation of intelligence✓ Link95.0GPT-6 Astra2026-09-03
ARC Prize Leaderboard94.1667GPT-6.1 Sol (Max)2026-09-29
ARC Prize Leaderboard✓ Link93.3333Claude Opus 5.5 (High)2026-09-22
ARC Prize Leaderboard✓ Link92.5GPT-5.6 Sol (Max)2026-07-09
ARC Prize Leaderboard✓ Link91.25Gemini 4 Argon (High)2026-09-30
Introducing Claude Opus 5✓ Link90.42Claude Opus 52026-07-24
ARC Prize Leaderboard90.0Claude Fable 5.1 (Max)2026-09-01
ARC Prize Leaderboard89.5833GPT-6 Sol (Max)2026-09-22
ARC Prize Leaderboard✓ Link89.167Claude Fable 5 (Max)2026-06-09
ARC Prize Leaderboard✓ Link89.1667Gemini 3.8 Flash (High)2026-09-02
Introducing GPT-5.585.0GPT-5.52026-04-23
ARC Prize Leaderboard84.58Gemini 3 Deep Think (2/26)2025-11-18
ARC Prize Leaderboard83.33GPT-5.4 Pro (XHigh)2026-03-05
Gemini 3.1 Pro: A smarter model for your most complex tasks77.1Gemini 3.1 Pro2026-02-19
ARC Prize Leaderboard75.83Claude 4.7 (Max)2026-04-16
ARC Prize Leaderboard✓ Link72.9167DeepSeek V4.1 Flash (Max)2026-09-17
ARC Prize Leaderboard72.08Gemini 3.5 Flash (High)2026-05-19
ARC Prize Leaderboard72.08Claude Opus 4.8 (High)2026-05-28
ARC Prize Leaderboard69.17Claude Opus 4.6 (120K, High)2026-02-05
ARC Prize Leaderboard67.0833Grok 4.6 (XHigh)2026-08-12
ARC Prize Leaderboard✓ Link65.8333GLM-5.3-Flash (Max)2026-08-26
ARC Prize Leaderboard✓ Link61.389DeepSeek V4 Flash 0731 (Max)2026-04-26
ARC Prize Leaderboard61.2500Claude Haiku 5.5 (XHigh)2026-10-07
ARC Prize Leaderboard60.42Claude Sonnet 4.6 (High)2026-02-17
ARC Prize Leaderboard60.417Gemini 3.6 Flash (High)2026-07-21
ARC Prize Leaderboard✓ Link60.417Kimi K3 (Max)2026-07-27
Introducing GPT-5.254.2GPT-5.2 Pro2025-12-11
ARC Prize Leaderboard52.64Grok 4.5 (High)2026-07-08
The Optimizer Is the Agent: Reasoning-Driven Search across Prompts, Programs, and ML Workflows✓ Link50.0ReASearch2026-08-07
Inkling-Small✓ Link40.1Inkling-Small2026-07-27
ARC Prize Leaderboard✓ Link36.528Inkling2026-07-15
Introducing Gemini 3 Flash33.6Google Gemini 3 Flash (Thinking)2025-12-17
ARC Prize Leaderboard27.64NVARC2026-01-15
ARC Prize Leaderboard✓ Link22.7750GLM-5.22026-06-16
Natural Image Pretraining Improves Abstract Reasoning13.1Nat-ARC ensemble2026-09-08
ARC Prize Leaderboard✓ Link11.81Kimi K2.52026-02-02
ARC Is a Vision Problem!11.1VARC (ensemble)2025-11-18
ARC Is a Vision Problem!8.3VARC2025-11-18
ARC Prize Leaderboard✓ Link6.53o3 (High)2025-04-16
ARC Prize Leaderboard✓ Link6.25Tiny Recursion Model (TRM)2025-10-06
Hierarchical Reasoning Model✓ Link5.0HRM2025-06-26
ARC Prize Leaderboard4.86Gemini 2.5 Pro (Thinking 32K)2025-07-07
ARC Prize Leaderboard✓ Link4.86MiniMax M2.52026-02-12
ARC Prize Leaderboard✓ Link4.03Deepseek V3.22025-12-02
ARC Prize Leaderboard2.5ARChitects2024-12-05
ARC Prize Leaderboard0.9Claude 3.7 (8K)2025-02-24
ARC Prize Leaderboard✓ Link0.0Llama 4 Maverick2025-04-05
ARC Prize Leaderboard0.0Magistral Medium (Thinking)2025-06-12