Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Logical Reasoning on ARC-AGI
Loading...
4.5
Pass@1
Random
-0.17896
1.03577
2.2505
3.46523
Jun 23, 2026
Pass@1
Updated 1mo ago
Evaluation Results
Method
Method
Links
Pass@1
Random
Backbone=Qwen3-4B-Base...
2026.06
4.5
SEC
Backbone=Qwen3-4B-Base...
2026.06
4.2
TAC
Backbone=Qwen3-4B-Base...
2026.06
2.6
TAC
Backbone=Qwen3-0.6B-Ba...
2026.06
0.5
Random
Backbone=Qwen3-0.6B-Ba...
2026.06
0.3
SEC
Backbone=Qwen3-0.6B-Ba...
2026.06
0.3
TAC
Backbone=Qwen3-1.7B
2026.06
0.014
TAC
Backbone=Llama3.2-3B
2026.06
0.012
Random
Backbone=Qwen3-1.7B
2026.06
0.01
Base
Backbone=Llama3.2-3B
2026.06
0.01
Random
Backbone=Llama3.2-3B
2026.06
0.008
M2O
Backbone=Qwen3-1.7B
2026.06
0.006
SEC
Backbone=Qwen3-1.7B
2026.06
0.005
M2O
Backbone=Llama3.2-3B
2026.06
0.004
Base
Backbone=Qwen3-1.7B
2026.06
0.001
SEC
Backbone=Llama3.2-3B
2026.06
0.001
Feedback
Search any
task
Search any
task