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 Germany. 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 | 3rd 67.1% | 6th 53.4% | 1st 68.5% | 2nd 68.0% | 5th 66.4% | 4th 66.7% | |
| 55.3% | 57% | 60% | 49% | 53% | 48% | 65% | |
| 72.5% | 69% | 67% | 85% | 73% | 76% | 65% | |
| 55.5% | 64% | 0% | 62% | 68% | 72% | 67% | |
| 59.5% | 58% | 52% | 57% | 63% | 75% | 52% | |
| 52.5% | 69% | 68% | 26% | 53% | 50% | 49% | |
| 61.7% | 73% | 0% | 76% | 78% | 70% | 73% | |
| 67.7% | 69% | 63% | 84% | 64% | 60% | 66% | |
| 69.3% | 74% | 63% | 83% | 65% | 60% | 71% | |
| 75.2% | 76% | 78% | 74% | 77% | 70% | 76% | |
| 81.0% | 62% | 83% | 89% | 86% | 83% | 83% |