Loading blueprint versions...
Please wait while we gather all the unique runs for this blueprint.
Please wait while we gather all the unique runs for this blueprint.
Please wait while we prepare the detailed comparison.
DTEF: Predict response distributions for United States. Source: Global Dialogues GD3 (https://github.com/collect-intel/global-dialogues)
Average key point coverage extent for each model across all prompts.
| Prompts vs. Models | Claude Haiku 4.5 | Gemini 2.5 Flash | Gemma 3 12b It | GPT 4.1 Mini | GPT 4.1 Nano | GPT 4o Mini | |
|---|---|---|---|---|---|---|---|
| Score | 2nd 83.1% | 1st 85.0% | 5th 79.7% | 3rd 82.9% | 6th 75.2% | 4th 79.8% | |
| 78.3% | 76% | 88% | 83% | 74% | 79% | 70% | |
| 83.8% | 91% | 92% | 82% | 94% | 62% | 82% | |
| 87.5% | 93% | 90% | 87% | 87% | 82% | 86% | |
| 76.0% | 77% | 76% | 81% | 80% | 68% | 74% | |
| 68.8% | 75% | 77% | 68% | 63% | 65% | 65% | |
| 83.8% | 77% | 90% | 88% | 83% | 84% | 81% | |
| 86.2% | 88% | 86% | 81% | 93% | 87% | 82% | |
| 82.7% | 85% | 85% | 77% | 93% | 67% | 89% | |
| 75.2% | 80% | 77% | 73% | 75% | 68% | 78% | |
| 87.2% | 89% | 89% | 77% | 87% | 90% | 91% |