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 South Africa. 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 Haiku 4.5 | Claude Sonnet 4 | Claude Sonnet 4.5 | Gemini 2.5 Flash | Gemma 3 12b It | Llama 4 Maverick | Mistral Large 2411 | Mistral Medium 3 | GPT 4.1 | GPT 4.1 Mini | GPT 4.1 Nano | GPT 4o | GPT 4o Mini | GPT 5 | GPT 5.1 | GPT OSS 120b | GPT OSS 20b | Qwen3 30b A3B Instruct 2507 | Qwen3 32b | Grok 4 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Score | 10th 75.0% | 6th 76.5% | 12th 74.5% | 5th 77.0% | 13th 74.3% | 20th 67.0% | 18th 69.8% | 19th 68.8% | 15th 73.5% | 2nd 79.3% | 13th 74.3% | 21st 66.8% | 3rd 77.5% | 8th 75.3% | 7th 76.3% | 16th 72.8% | 11th 74.5% | 17th 71.7% | 1st 82.5% | 8th 75.3% | 4th 77.0% | |
| 69.4% | 71% | 74% | 60% | 69% | 68% | 54% | 65% | 67% | 64% | 78% | 77% | 60% | 83% | 82% | 78% | 67% | 65% | 61% | 75% | 77% | 63% | |
| 77.6% | 78% | 77% | 84% | 84% | 75% | 76% | 80% | 77% | 76% | 82% | 77% | 78% | 85% | 77% | 79% | 73% | 68% | 73% | 85% | 68% | 77% | |
| 66.3% | 71% | 69% | 69% | 71% | 82% | 59% | 51% | 51% | 74% | 74% | 59% | 54% | 64% | 59% | 65% | 71% | 68% | 69% | 78% | 61% | 73% | |
| 83.7% | 80% | 86% | 85% | 84% | 72% | 79% | 83% | 80% | 80% | 83% | 84% | 75% | 78% | 83% | 83% | 80% | 97% | 84% | 92% | 95% | 95% |