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 | 2nd 78.3% | 10th 76.3% | 9th 76.8% | 6th 77.3% | 18th 70.5% | 21st 65.8% | 19th 69.8% | 20th 68.8% | 15th 73.3% | 4th 77.8% | 12th 75.0% | 16th 73.0% | 8th 77.0% | 13th 74.0% | 5th 77.8% | 11th 75.3% | 17th 72.0% | 13th 74.0% | 3rd 78.0% | 6th 77.3% | 1st 88.5% | |
| 68.3% | 71% | 68% | 60% | 69% | 67% | 54% | 65% | 67% | 61% | 69% | 72% | 77% | 78% | 82% | 75% | 70% | 66% | 62% | 66% | 59% | 77% | |
| 80.2% | 84% | 78% | 84% | 85% | 77% | 71% | 80% | 77% | 77% | 85% | 82% | 84% | 84% | 77% | 82% | 79% | 72% | 73% | 85% | 81% | 87% | |
| 67.9% | 74% | 74% | 78% | 71% | 71% | 59% | 51% | 51% | 71% | 74% | 62% | 57% | 54% | 59% | 71% | 69% | 66% | 77% | 69% | 72% | 95% | |
| 83.8% | 84% | 85% | 85% | 84% | 67% | 79% | 83% | 80% | 84% | 83% | 84% | 74% | 92% | 78% | 83% | 83% | 84% | 84% | 92% | 97% | 95% |