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 Indonesia. Source: Global Dialogues GD3 (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 75.3% | 1st 77.8% | 4th 71.1% | 5th 69.9% | 6th 66.9% | 3rd 73.5% | |
| 75.5% | 83% | 89% | 77% | 72% | 60% | 72% | |
| 67.2% | 78% | 84% | 55% | 47% | 61% | 78% | |
| 77.3% | 81% | 80% | 71% | 73% | 80% | 79% | |
| 75.7% | 75% | 80% | 79% | 78% | 62% | 80% | |
| 48.8% | 44% | 64% | 46% | 54% | 39% | 46% | |
| 78.3% | 72% | 84% | 82% | 71% | 85% | 76% | |
| 78.7% | 83% | 86% | 79% | 77% | 72% | 75% | |
| 75.2% | 74% | 68% | 79% | 89% | 75% | 66% | |
| 69.3% | 77% | 69% | 79% | 52% | 65% | 74% | |
| 78.2% | 86% | 74% | 64% | 86% | 70% | 89% |