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 United States. 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 | 19th 76.8% | 2nd 84.3% | 9th 81.5% | 20th 76.3% | 15th 79.0% | 16th 78.0% | 5th 82.8% | 11th 80.0% | 10th 81.0% | 13th 79.5% | 1st 85.5% | 11th 80.0% | 6th 82.5% | 3rd 83.3% | 8th 81.5% | 3rd 83.3% | 16th 78.0% | 18th 77.3% | 21st 70.5% | 7th 82.0% | 14th 79.3% | |
| 83.8% | 81% | 94% | 83% | 73% | 81% | 77% | 91% | 81% | 89% | 87% | 95% | 80% | 85% | 89% | 91% | 96% | 87% | 90% | 40% | 83% | 87% | |
| 86.7% | 78% | 93% | 91% | 85% | 83% | 89% | 86% | 90% | 81% | 77% | 87% | 86% | 93% | 90% | 90% | 82% | 88% | 91% | 85% | 89% | 87% | |
| 66.6% | 70% | 69% | 70% | 70% | 69% | 68% | 72% | 62% | 60% | 63% | 71% | 69% | 69% | 69% | 62% | 62% | 61% | 56% | 70% | 69% | 67% | |
| 83.3% | 78% | 81% | 82% | 77% | 83% | 78% | 82% | 87% | 94% | 91% | 89% | 85% | 83% | 85% | 83% | 93% | 76% | 72% | 87% | 87% | 76% |