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 26-35. 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 64.9% | 5th 60.1% | 2nd 72.6% | 4th 64.5% | 6th 48.6% | 1st 81.4% | |
| 68.7% | 56% | 75% | 80% | 79% | 40% | 82% | |
| 81.0% | 85% | 92% | 83% | 78% | 68% | 80% | |
| 62.0% | 62% | 45% | 83% | 49% | 43% | 90% | |
| 58.5% | 64% | 44% | 55% | 55% | 46% | 87% | |
| 68.7% | 66% | 60% | 81% | 66% | 48% | 91% | |
| 69.0% | 69% | 67% | 87% | 72% | 46% | 73% | |
| 29.2% | 19% | 19% | 29% | 30% | 30% | 48% | |
| 72.5% | 65% | 62% | 78% | 82% | 66% | 82% | |
| 79.3% | 93% | 69% | 90% | 74% | 55% | 95% | |
| 64.7% | 70% | 68% | 60% | 60% | 44% | 86% |