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 59.5% | 5th 54.7% | 6th 48.0% | 2nd 59.0% | 4th 58.0% | 3rd 58.0% | |
| 72.0% | 84% | 64% | 63% | 70% | 78% | 73% | |
| 66.2% | 72% | 60% | 53% | 76% | 64% | 72% | |
| 55.3% | 57% | 52% | 43% | 69% | 60% | 51% | |
| 72.7% | 79% | 75% | 62% | 72% | 78% | 70% | |
| 74.3% | 73% | 77% | 60% | 87% | 75% | 74% | |
| 46.5% | 49% | 47% | 40% | 47% | 46% | 50% | |
| 56.2% | 50% | 43% | 68% | 51% | 59% | 66% | |
| 34.5% | 46% | 32% | 18% | 37% | 39% | 35% | |
| 47.0% | 42% | 61% | 39% | 47% | 38% | 55% | |
| 37.3% | 43% | 36% | 34% | 34% | 43% | 34% |