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 Judaism. 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 | 3rd 73.4% | 1st 74.3% | - - | 2nd 74.0% | 5th 63.7% | 4th 70.1% | |
| 70.6% | 68% | 78% | 67% | 77% | 63% | ||
| 62.0% | 69% | 70% | 68% | 46% | 57% | ||
| 82.4% | 84% | 86% | 79% | 80% | 83% | ||
| 84.6% | 84% | 85% | 86% | 86% | 82% | ||
| 53.6% | 53% | 64% | 47% | 40% | 64% | ||
| 71.8% | 72% | 72% | 78% | 65% | 72% | ||
| 76.8% | 80% | 76% | 75% | 74% | 79% | ||
| 64.0% | 74% | 66% | 75% | 54% | 51% | ||
| 79.2% | 83% | 88% | 90% | 55% | 80% | ||
| 66.0% | 67% | 58% | 75% | 60% | 70% |