Open-ended Question Answering on PopQA
50.8AccuracyGemini
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| GeminiProvider=Gemini, Sample Size (N)=19932026.05 | 50.8 | 0.4585 | 0.313 | 0.3069 | |
| AnthropicProvider=Anthropic, Sample Size (N)=4992026.05 | 50.5 | 0.2765 | 0.3968 | 0.3893 | |
| xAIProvider=xAI, Sample Size (N)=18962026.05 | 40.9 | 0.4308 | 0.3133 | 0.3016 | |
| OpenAIProvider=OpenAI, Sample Size (N)=19972026.05 | 39.9 | 0.3759 | 0.2721 | 0.2612 | |
| MistralProvider=Mistral, Sample Size (N)=20002026.05 | 38.9 | 0.4722 | 0.4795 | 0.4735 |