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 Korea South. Source: Global Dialogues GD7 (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 | 4th 69.3% | 3rd 70.4% | 6th 61.7% | 2nd 71.8% | 1st 74.0% | 5th 67.2% | |
| 77.8% | 81% | 77% | 89% | 73% | 75% | 72% | |
| 71.2% | 76% | 77% | 52% | 75% | 73% | 74% | |
| 71.3% | 76% | 66% | 66% | 70% | 75% | 75% | |
| 87.7% | 90% | 89% | 84% | 90% | 95% | 78% | |
| 60.2% | 65% | 65% | 47% | 61% | 66% | 57% | |
| 60.8% | 66% | 65% | 46% | 64% | 67% | 57% | |
| 56.7% | 60% | 63% | 44% | 60% | 58% | 55% | |
| 67.2% | 67% | 66% | 49% | 75% | 74% | 72% | |
| 79.5% | 69% | 81% | 76% | 92% | 81% | 78% | |
| 58.3% | 43% | 55% | 64% | 58% | 76% | 54% |