Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Error recognition on GAIA
Loading...
46
Tolerant Accuracy (±1 step)
CORRECT
35.6
38.3
41
43.7
Sep 28, 2025
Tolerant Accuracy (±1 step)
Improvement
Updated 1mo ago
Evaluation Results
Method
Method
Links
Tolerant Accuracy (±1 step)
Improvement
CORRECT
Base Model=GPT-5-nano,...
2025.09
46
10
Baseline
Base Model=GPT-5-nano,...
2025.09
36
-
Feedback
Search any
task
Search any
task
