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 Kingdom. 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 | 14th 76.3% | 9th 79.8% | 10th 79.3% | 18th 73.3% | 17th 74.5% | 21st 70.0% | 7th 81.8% | 8th 80.3% | 13th 76.8% | 20th 71.0% | 6th 82.0% | 15th 76.0% | 4th 82.5% | 19th 73.0% | 4th 82.5% | 11th 78.5% | 12th 78.0% | 16th 75.8% | 1st 84.0% | 2nd 83.2% | 3rd 83.0% | |
| 74.3% | 74% | 83% | 70% | 63% | 73% | 56% | 81% | 67% | 75% | 64% | 82% | 65% | 87% | 55% | 84% | 82% | 76% | 79% | 84% | 81% | 79% | |
| 80.1% | 68% | 73% | 86% | 72% | 62% | 84% | 85% | 87% | 76% | 73% | 83% | 79% | 86% | 86% | 86% | 72% | 82% | 86% | 86% | 86% | 85% | |
| 71.9% | 78% | 76% | 82% | 83% | 74% | 63% | 76% | 74% | 65% | 64% | 82% | 76% | 70% | 60% | 73% | 67% | 64% | 53% | 72% | 76% | 81% | |
| 86.3% | 85% | 87% | 79% | 75% | 89% | 77% | 85% | 93% | 91% | 83% | 81% | 84% | 87% | 91% | 87% | 93% | 90% | 85% | 94% | 90% | 87% |