Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Abstract Reasoning on ARC-AGI 2 (Semi-private)
Loading...
100
Accuracy
Human Panel
30.944
48.872
66.8
84.728
Jun 30, 2026
Accuracy
Cost per Task
Updated 24d ago
Evaluation Results
Method
Method
Links
Accuracy
Cost per Task
Human Panel
Author=Human, Comment=...
2026.06
100
17
Modality-Driven Search + Holistic Trace Judging
Author=Johan Land, Com...
2026.06
72.9
38.99
GPT-5.2 Pro (High)
Author=OpenAI
2026.06
54.2
15.72
Gemini 3 Pro (Refine.)
Author=Poetiq
2026.06
54
30.57
GPT-5.2 (X-High)
Author=OpenAI
2026.06
52.9
1.9
Gemini 3 Deep Think (Preview)
Author=Google
2026.06
45.1
77.16
GPT-5.2 (High)
Author=OpenAI
2026.06
43.3
1.39
GPT-5.2 Pro (Medium)
Author=OpenAI
2026.06
38.5
8.99
Opus 4.5 (Thinking, 64K)
Author=Anthropic
2026.06
37.6
2.4
Gemini 3 Flash Preview (High)
Author=Google
2026.06
33.6
0.23
Feedback
Search any
task
Search any
task