List your company
ResearchFormerly WizWand

Biology Laboratory Reasoning on LabBench

74.4AccuracyGPT-5
52.5658.2363.969.57Aug 26, 2025
Updated 5mo ago

Evaluation results

MethodLinks
GPT-5Reasoning Effort=High2025.0874.47.8
o3Reasoning Effort=High2025.0874.23.7
o3Reasoning Effort=Low2025.0870.5—
GPT-5Reasoning Effort=Low2025.0866.6—
Gemini-2.5-ProReasoning Effort=High2025.0864.42.5
o4-miniReasoning Effort=High2025.0863.74
Gemini-2.5-ProReasoning Effort=Low2025.0861.9—
o4-miniReasoning Effort=Low2025.0859.7—
o3-miniReasoning Effort=High2025.0859.22.3
Claude-Sonnet-4Reasoning Effort=High2025.0857.23.8
o3-miniReasoning Effort=Low2025.0856.9—
Claude-Sonnet-4Reasoning Effort=Low2025.0853.4—