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 Equally concerned and excited. 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 | 2nd 78.4% | 1st 82.4% | 3rd 78.2% | 4th 77.6% | 5th 72.0% | 6th 70.2% | |
| 58.5% | 72% | 77% | 48% | 55% | 48% | 51% | |
| 78.5% | 77% | 83% | 85% | 82% | 72% | 72% | |
| 71.0% | 71% | 77% | 79% | 65% | 69% | 65% | |
| 68.5% | 69% | 76% | 66% | 74% | 69% | 57% | |
| 87.5% | 88% | 86% | 86% | 89% | 86% | 90% | |
| 74.3% | 78% | 69% | 82% | 80% | 67% | 70% | |
| 88.7% | 94% | 96% | 90% | 86% | 83% | 83% | |
| 85.6% | 88% | 87% | 88% | 87% | 78% | ||
| 72.0% | 69% | 89% | 77% | 70% | 65% | 62% | |
| 81.3% | 78% | 84% | 91% | 87% | 74% | 74% |