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 GD5 (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 61.4% | 1st 64.7% | 4th 59.3% | 2nd 63.1% | 6th 53.8% | 5th 57.9% | |
| 61.2% | 61% | 54% | 71% | 57% | 63% | ||
| 73.2% | 73% | 70% | 67% | 85% | 70% | 74% | |
| 71.2% | 62% | 72% | 80% | 64% | 70% | 79% | |
| 72.8% | 84% | 82% | 77% | 92% | 55% | 47% | |
| 54.3% | 59% | 47% | 55% | 59% | 45% | 61% | |
| 72.0% | 69% | 64% | 74% | 83% | 67% | 75% | |
| 46.0% | 50% | 58% | 47% | 37% | 47% | 37% | |
| 61.5% | 63% | 63% | 53% | 58% | 65% | 67% | |
| 35.3% | 35% | 63% | 29% | 36% | 17% | 32% | |
| 52.2% | 58% | 63% | 57% | 46% | 45% | 44% |