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 | 4th 91.3% | 9th 90.3% | 2nd 92.5% | 12th 89.5% | 14th 88.8% | 20th 83.0% | 10th 90.0% | 17th 86.5% | 18th 85.0% | 18th 85.0% | 4th 91.3% | 7th 90.8% | 6th 91.0% | 21st 76.3% | 1st 93.0% | 15th 88.3% | 12th 89.5% | 16th 88.0% | 11th 89.7% | 8th 90.5% | 3rd 91.7% | |
| 88.9% | 98% | 96% | 98% | 87% | 94% | 70% | 86% | 82% | 89% | 83% | 96% | 88% | 91% | 67% | 90% | 95% | 90% | 88% | 93% | 89% | 96% | |
| 89.0% | 90% | 86% | 92% | 91% | 88% | 86% | 89% | 90% | 86% | 77% | 92% | 91% | 92% | 83% | 97% | 87% | 97% | 89% | 87% | 90% | 88% | |
| 85.5% | 90% | 88% | 89% | 89% | 87% | 88% | 89% | 86% | 78% | 83% | 91% | 86% | 85% | 67% | 87% | 85% | 80% | 83% | 89% | 91% | ||
| 91.0% | 87% | 91% | 91% | 91% | 86% | 88% | 96% | 88% | 87% | 97% | 86% | 98% | 96% | 88% | 98% | 86% | 91% | 92% | 92% | 91% |