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 [synth]: Predict response distributions for Judaism. 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 | 5th 54.0% | 2nd 56.7% | 3rd 55.8% | 3rd 55.8% | 6th 32.8% | 1st 69.5% | |
| 58.0% | 58% | 57% | 80% | 69% | 17% | 67% | |
| 75.5% | 82% | 88% | 81% | 82% | 34% | 86% | |
| 54.2% | 50% | 57% | 67% | 60% | 15% | 76% | |
| 40.0% | 43% | 22% | 51% | 26% | 25% | 73% | |
| 66.7% | 60% | 68% | 56% | 73% | 69% | 74% | |
| 67.5% | 69% | 69% | 66% | 68% | 59% | 74% | |
| 14.8% | 0% | 59% | 7% | 0% | 6% | 17% | |
| 59.0% | 54% | 67% | 59% | 63% | 29% | 82% | |
| 44.7% | 59% | 21% | 36% | 59% | 15% | 78% | |
| 60.7% | 65% | 59% | 55% | 58% | 59% | 68% |