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 Italy. Source: Global Dialogues GD1 (https://github.com/collect-intel/global-dialogues)
Average key point coverage extent for each model across all prompts.
| Prompts vs. Models | Claude 3.7 Sonnet | Claude 3.5 Haiku | Claude Haiku 4.5 | Claude Sonnet 4 | Gemini 2.5 Flash | Gemma 3 12b It | Llama 3 70b Instruct | Llama 4 Maverick | Mistral Large 2411 | Mistral Medium 3 | Mistral Nemo | GPT 4.1 | GPT 4.1 Mini | GPT 4.1 Nano | GPT 4o | GPT 4o Mini | GPT OSS 120b | GPT OSS 20b | GLM 4.5 | Qwen3 30b A3B Instruct 2507 | Qwen3 32b | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Score | 20th 69.3% | 5th 77.8% | 10th 76.0% | 19th 70.5% | 18th 70.8% | 16th 72.0% | 11th 75.3% | 17th 71.3% | 14th 72.3% | 13th 72.5% | 3rd 78.5% | 14th 72.3% | 2nd 79.5% | 21st 68.8% | 12th 74.5% | 6th 77.5% | 1st 80.8% | 8th 76.3% | 7th 76.8% | 8th 76.3% | 3rd 78.5% | |
| 73.2% | 67% | 80% | 67% | 57% | 63% | 57% | 76% | 60% | 73% | 70% | 86% | 60% | 83% | 69% | 72% | 82% | 90% | 82% | 82% | 72% | 90% | |
| 84.0% | 68% | 83% | 90% | 83% | 77% | 91% | 86% | 81% | 76% | 75% | 85% | 78% | 87% | 76% | 87% | 90% | 89% | 96% | 87% | 87% | 92% | |
| 60.6% | 65% | 65% | 70% | 65% | 60% | 63% | 66% | 56% | 57% | 62% | 72% | 62% | 65% | 47% | 56% | 55% | 61% | 46% | 61% | 63% | 56% | |
| 80.6% | 77% | 83% | 77% | 77% | 83% | 77% | 73% | 88% | 83% | 83% | 71% | 89% | 83% | 83% | 83% | 83% | 83% | 81% | 77% | 83% | 76% |