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 Japan. 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 | 7th 71.3% | 10th 69.0% | 12th 65.7% | 11th 68.7% | 13th 63.0% | 5th 72.7% | 6th 72.3% | 8th 71.0% | 3rd 76.3% | 9th 70.0% | 2nd 77.0% | 1st 80.0% | 4th 73.7% | |
| 84.7% | 82% | 80% | 80% | 94% | 80% | 89% | 84% | 89% | 84% | 89% | 87% | 93% | 70% | |
| 74.5% | 72% | 78% | 68% | 69% | 65% | 71% | 70% | 75% | 78% | 64% | 77% | 91% | 91% | |
| 55.5% | 60% | 49% | 49% | 43% | 44% | 58% | 63% | 49% | 67% | 57% | 67% | 56% | 60% |