List your company

Open-ended Question Answering on PopQA

50.8AccuracyGemini
38.42441.63744.8548.063May 8, 2026

Evaluation results

MethodLinks
GeminiProvider=Gemini, Sample Size (N)=19932026.0550.80.45850.3130.3069
AnthropicProvider=Anthropic, Sample Size (N)=4992026.0550.50.27650.39680.3893
xAIProvider=xAI, Sample Size (N)=18962026.0540.90.43080.31330.3016
OpenAIProvider=OpenAI, Sample Size (N)=19972026.0539.90.37590.27210.2612
MistralProvider=Mistral, Sample Size (N)=20002026.0538.90.47220.47950.4735