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 South Africa. 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 | 3rd 60.7% | 2nd 62.2% | 6th 49.1% | 4th 59.3% | 1st 63.4% | 5th 56.4% | |
| 75.1% | 82% | 74% | 58% | 79% | 87% | 71% | |
| 69.3% | 73% | 74% | 53% | 81% | 60% | 76% | |
| 59.4% | 54% | 64% | 63% | 59% | 64% | 53% | |
| 66.9% | 66% | 71% | 69% | 71% | 71% | 54% | |
| 67.3% | 76% | 71% | 64% | 64% | 64% | 65% | |
| 49.5% | 46% | 52% | 36% | 47% | 69% | 47% | |
| 72.0% | 72% | 78% | 70% | 72% | 70% | 71% | |
| 40.4% | 48% | 44% | 20% | 40% | 52% | 40% | |
| 47.2% | 52% | 52% | 33% | 46% | 45% | 55% | |
| 38.2% | 38% | 44% | 25% | 36% | 54% | 33% |