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 Islam. Source: Global Dialogues GD4 (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 80.5% | 5th 75.6% | 4th 75.9% | 1st 82.8% | 3rd 79.0% | 6th 74.8% | |
| 85.8% | 91% | 88% | 78% | 87% | 87% | 84% | |
| 66.0% | 68% | 64% | 60% | 85% | 59% | 60% | |
| 84.5% | 87% | 84% | 84% | 86% | 84% | 82% | |
| 64.0% | 61% | 47% | 80% | 63% | 74% | 59% | |
| 81.7% | 87% | 87% | 73% | 93% | 80% | 70% | |
| 83.8% | 85% | 83% | 87% | 84% | 78% | 86% | |
| 75.3% | 70% | 50% | 81% | 90% | 86% | 75% | |
| 83.8% | 91% | 89% | 80% | 84% | 79% | 80% | |
| 72.7% | 74% | 74% | 66% | 74% | 78% | 70% | |
| 83.3% | 91% | 90% | 70% | 82% | 85% | 82% |