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 Malaysia. Source: Global Dialogues GD5 (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 66.3% | 6th 61.9% | 5th 63.3% | 3rd 64.6% | 4th 63.4% | 1st 68.8% | |
| 62.7% | 66% | 63% | 60% | 63% | 63% | 61% | |
| 79.7% | 76% | 78% | 86% | 78% | 77% | 83% | |
| 52.0% | 45% | 45% | 61% | 50% | 50% | 61% | |
| 73.0% | 75% | 78% | 66% | 75% | 68% | 76% | |
| 55.5% | 61% | 49% | 55% | 54% | 57% | 57% | |
| 56.8% | 60% | 70% | 59% | 63% | 30% | 59% | |
| 76.5% | 75% | 74% | 58% | 72% | 92% | 88% | |
| 60.5% | 65% | 56% | 53% | 60% | 68% | 61% | |
| 55.7% | 66% | 39% | 66% | 51% | 46% | 66% | |
| 74.8% | 74% | 67% | 69% | 80% | 83% | 76% |