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 | 1st 58.1% | 5th 54.0% | 6th 48.4% | 4th 55.9% | 3rd 56.5% | 2nd 58.1% | |
| 70.3% | 82% | 64% | 63% | 70% | 70% | 73% | |
| 65.3% | 62% | 58% | 55% | 74% | 74% | 69% | |
| 56.2% | 55% | 52% | 46% | 72% | 61% | 51% | |
| 64.2% | 73% | 78% | 55% | 72% | 37% | 70% | |
| 72.8% | 77% | 78% | 55% | 74% | 77% | 76% | |
| 48.3% | 50% | 48% | 40% | 45% | 59% | 48% | |
| 57.8% | 50% | 43% | 67% | 47% | 67% | 73% | |
| 35.5% | 44% | 34% | 21% | 37% | 37% | 40% | |
| 47.0% | 45% | 56% | 51% | 41% | 42% | 47% | |
| 34.2% | 43% | 29% | 31% | 27% | 41% | 34% |