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 Judaism. 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 62.8% | 1st 64.0% | 4th 57.5% | 3rd 60.5% | 6th 52.0% | 5th 57.3% | |
| 61.5% | 66% | 54% | 61% | 72% | 58% | 58% | |
| 68.8% | 72% | 70% | 73% | 73% | 51% | 74% | |
| 67.7% | 70% | 70% | 81% | 62% | 58% | 65% | |
| 74.3% | 78% | 92% | 62% | 83% | 75% | 56% | |
| 56.7% | 63% | 50% | 57% | 59% | 50% | 61% | |
| 65.8% | 76% | 58% | 67% | 76% | 43% | 75% | |
| 43.0% | 43% | 67% | 33% | 37% | 41% | 37% | |
| 62.7% | 61% | 67% | 55% | 61% | 61% | 71% | |
| 36.8% | 33% | 59% | 29% | 32% | 36% | 32% | |
| 52.8% | 66% | 53% | 57% | 50% | 47% | 44% |