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 Christianity. Source: Global Dialogues GD7 (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 | 3rd 83.3% | 2nd 85.7% | 6th 78.1% | 1st 87.5% | 4th 82.5% | 5th 81.3% | |
| 91.5% | 92% | 91% | 92% | 92% | 91% | 91% | |
| 78.2% | 81% | 85% | 62% | 87% | 84% | 70% | |
| 69.3% | 65% | 79% | 63% | 67% | 75% | 67% | |
| 87.0% | 90% | 89% | 81% | 90% | 81% | 91% | |
| 90.2% | 87% | 89% | 85% | 97% | 93% | ||
| 85.5% | 84% | 89% | 78% | 97% | 87% | 78% | |
| 81.5% | 84% | 89% | 79% | 80% | 70% | 87% | |
| 79.8% | 82% | 81% | 72% | 81% | 81% | 82% | |
| 82.5% | 88% | 81% | 80% | 90% | 75% | 81% | |
| 86.7% | 80% | 84% | 89% | 94% | 88% | 85% |