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 | 1st 54.4% | 5th 47.9% | 3rd 49.3% | 2nd 50.9% | 6th 41.9% | 4th 48.6% | |
| 41.7% | 43% | 38% | 65% | 43% | 29% | 32% | |
| 95.2% | 71% | 100% | 100% | 100% | 100% | 100% | |
| 39.3% | 40% | 34% | 41% | 41% | 41% | 39% | |
| 49.2% | 64% | 49% | 49% | 49% | 35% | 49% | |
| 58.5% | 60% | 60% | 54% | 59% | 58% | 60% | |
| 38.3% | 59% | 35% | 53% | 38% | 21% | 24% | |
| 26.7% | 25% | 26% | 27% | 26% | 20% | 36% | |
| 49.0% | 60% | 45% | 44% | 60% | 42% | 43% | |
| 46.5% | 77% | 46% | 0% | 46% | 46% | 64% | |
| 44.0% | 45% | 46% | 60% | 47% | 27% | 39% |