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 | 4th 64.4% | 6th 47.1% | 5th 57.7% | 3rd 64.6% | 2nd 64.9% | 1st 67.3% | |
| 54.2% | 71% | 61% | 10% | 66% | 59% | 58% | |
| 66.8% | 83% | 0% | 83% | 70% | 77% | 88% | |
| 51.0% | 44% | 45% | 60% | 43% | 54% | 60% | |
| 68.3% | 77% | 75% | 57% | 71% | 59% | 71% | |
| 55.8% | 53% | 48% | 60% | 56% | 61% | 57% | |
| 49.0% | 60% | 0% | 67% | 58% | 58% | 51% | |
| 76.3% | 70% | 69% | 67% | 82% | 87% | 83% | |
| 58.2% | 61% | 58% | 53% | 58% | 62% | 57% | |
| 53.0% | 51% | 40% | 58% | 55% | 52% | 62% | |
| 77.3% | 74% | 75% | 62% | 87% | 80% | 86% |