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 GD6 (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 85.0% | 5th 80.6% | 5th 80.6% | 1st 85.2% | 4th 82.6% | 3rd 83.3% | |
| 80.3% | 80% | 81% | 78% | 77% | 85% | 81% | |
| 82.3% | 80% | 76% | 82% | 86% | 81% | 89% | |
| 84.2% | 88% | 90% | 84% | 86% | 72% | 85% | |
| 65.0% | 61% | 43% | 69% | 83% | 66% | 68% | |
| 83.5% | 93% | 86% | 70% | 75% | 87% | 90% | |
| 87.2% | 88% | 90% | 87% | 92% | 83% | 83% | |
| 86.8% | 89% | 90% | 85% | 85% | 86% | 86% | |
| 84.7% | 89% | 79% | 79% | 92% | 87% | 82% | |
| 86.2% | 91% | 90% | 75% | 86% | 89% | 86% | |
| 88.7% | 91% | 81% | 97% | 90% | 90% | 83% |