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 Islam. Source: Global Dialogues GD2 (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 | Claude Sonnet 4 | Claude Sonnet 4.5 | Gemini 2.5 Flash | Gemma 3 12b It | Llama 4 Maverick | Mistral Medium 3 | GPT 4.1 | GPT 4.1 Mini | GPT 4.1 Nano | GPT 4o Mini | Qwen3 30b A3B Instruct 2507 | Qwen3 Next 80b A3B Instruct | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Score | 1st 85.7% | 5th 81.7% | 7th 81.3% | 12th 77.0% | 11th 77.3% | 10th 79.0% | 9th 79.3% | 2nd 84.3% | 4th 82.3% | 8th 81.0% | 3rd 83.0% | 13th 58.7% | 5th 81.7% | |
| 89.2% | 94% | 90% | 91% | 75% | 91% | 88% | 91% | 91% | 91% | 91% | 88% | 88% | 90% | |
| 73.3% | 86% | 86% | 79% | 82% | 77% | 72% | 76% | 82% | 76% | 72% | 76% | 10% | 79% | |
| 75.8% | 77% | 69% | 74% | 74% | 64% | 77% | 71% | 80% | 80% | 80% | 85% | 78% | 76% |