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 [synth]: Predict response distributions for I do not identify with any religious group or faith. 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 | 2nd 67.2% | 5th 49.0% | 4th 61.8% | 3rd 66.4% | 6th 48.4% | 1st 83.1% | |
| 64.2% | 65% | 65% | 60% | 66% | 47% | 82% | |
| 69.3% | 90% | 43% | 43% | 85% | 68% | 87% | |
| 62.7% | 72% | 54% | 73% | 61% | 27% | 89% | |
| 51.3% | 44% | 41% | 59% | 48% | 33% | 83% | |
| 63.0% | 68% | 66% | 50% | 60% | 50% | 84% | |
| 77.0% | 82% | 82% | 77% | 81% | 63% | 77% | |
| 35.3% | 25% | 8% | 32% | 25% | 56% | 66% | |
| 67.5% | 72% | 59% | 61% | 89% | 41% | 83% | |
| 78.3% | 84% | 72% | 95% | 77% | 52% | 90% | |
| 57.8% | 70% | 0% | 68% | 72% | 47% | 90% |