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 More excited than concerned. 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 | 3rd 55.2% | 4th 47.2% | 1st 58.4% | 2nd 57.0% | 6th 41.9% | 5th 47.2% | |
| 44.3% | 43% | 42% | 64% | 47% | 30% | 40% | |
| 90.7% | 71% | 100% | 100% | 100% | 100% | 73% | |
| 40.3% | 44% | 35% | 41% | 41% | 41% | 40% | |
| 53.3% | 64% | 45% | 49% | 66% | 35% | 61% | |
| 59.7% | 60% | 60% | 59% | 59% | 60% | 60% | |
| 39.5% | 58% | 35% | 57% | 37% | 18% | 32% | |
| 24.8% | 25% | 23% | 27% | 25% | 20% | 29% | |
| 50.5% | 62% | 45% | 45% | 62% | 42% | 47% | |
| 63.2% | 79% | 46% | 75% | 87% | 46% | 46% | |
| 45.2% | 46% | 41% | 67% | 46% | 27% | 44% |