Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Scientific Question Answering on GPQA (Accuracy (%), Δ)
Loading...
82.4
Accuracy
GPT-5
62.64
67.77
72.9
78.03
Aug 26, 2025
Accuracy
Delta (Δ)
Updated 5d ago
Evaluation Results
Method
Method
Links
Accuracy
Delta (Δ)
GPT-5
Reasoning Effort=High
2025.08
82.4
3.1
Gemini-2.5-Pro
Reasoning Effort=Low
2025.08
80.1
-
o3
Reasoning Effort=High
2025.08
79.9
4.5
Gemini-2.5-Pro
Reasoning Effort=High
2025.08
79.5
-0.6
GPT-5
Reasoning Effort=Low
2025.08
79.2
-
o3
Reasoning Effort=Low
2025.08
75.4
-
o4-mini
Reasoning Effort=High
2025.08
74.6
5.2
o3-mini
Reasoning Effort=High
2025.08
73.9
10.5
o4-mini
Reasoning Effort=Low
2025.08
69.4
-
Claude-Sonnet-4
Reasoning Effort=High
2025.08
69
5.2
Claude-Sonnet-4
Reasoning Effort=Low
2025.08
63.8
-
o3-mini
Reasoning Effort=Low
2025.08
63.4
-
Feedback
Search any
task
Search any
task