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 China. 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 | 3rd 92.8% | 5th 91.5% | 9th 90.8% | 13th 88.5% | 10th 90.5% | 21st 80.0% | 7th 91.0% | 17th 86.0% | 19th 81.3% | 18th 85.0% | 2nd 93.2% | 6th 91.3% | 4th 91.8% | 20th 81.0% | 1st 93.3% | 16th 86.8% | 14th 88.3% | 7th 91.0% | 11th 89.8% | 12th 89.0% | 15th 87.3% | |
| 89.2% | 99% | 96% | 90% | 87% | 94% | 63% | 92% | 82% | 85% | 83% | 96% | 88% | 98% | 73% | 92% | 95% | 90% | 96% | 96% | 82% | 96% | |
| 87.7% | 94% | 91% | 92% | 91% | 88% | 86% | 87% | 85% | 79% | 77% | 92% | 91% | 87% | 87% | 97% | 85% | 86% | 88% | 84% | 89% | 85% | |
| 85.9% | 89% | 88% | 90% | 89% | 90% | 86% | 89% | 86% | 78% | 83% | 88% | 89% | 86% | 78% | 86% | 79% | 79% | 89% | 85% | 92% | 84% | |
| 91.5% | 89% | 91% | 91% | 87% | 90% | 85% | 96% | 91% | 83% | 97% | 97% | 97% | 96% | 86% | 98% | 88% | 98% | 91% | 94% | 93% | 84% |