Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Scientific Reasoning on OlympiadBench Physics
Loading...
87.3
Accuracy
HEART
5.6392
26.8396
48.04
69.2404
Sep 26, 2025
Accuracy
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
HEART
Model=Deepseek-Reasoner
2025.09
87.3
HEART
Model=Claude 4 Sonnet
2025.09
86.67
HEART
Model=Gemini 2.5 Pro
2025.09
84.44
HEART
Model=Gemini 2.5 Flash
2025.09
84.13
HEART
Model=GPT-5 nano
2025.09
83.81
Vanilla
Model=Claude 4 Sonnet
2025.09
67.94
Vanilla
Model=Gemini 2.5 Flash
2025.09
65.82
Vanilla
Model=GPT-5 nano
2025.09
62.86
Vanilla
Model=Deepseek-Reasoner
2025.09
60.63
Vanilla
Model=Gemini 2.5 Pro
2025.09
60.11
HEART
Model=Gemma3 12b Instruct
2025.09
53.02
HEART
Model=Gemma3 4b Instruct
2025.09
25.19
Vanilla
Model=Gemma3 12b Instruct
2025.09
22.96
Vanilla
Model=Gemma3 4b Instruct
2025.09
8.78
Feedback
Search any
task
Search any
task