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 GD2 (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 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Score | 3rd 66.3% | 5th 65.0% | 7th 62.7% | 5th 65.0% | 2nd 68.3% | 20th 51.7% | 19th 53.3% | 18th 53.7% | 15th 58.3% | 8th 62.3% | 4th 65.3% | 12th 59.3% | 15th 58.3% | 10th 60.3% | 1st 68.7% | 8th 62.3% | 17th 57.7% | 10th 60.3% | 14th 58.7% | 13th 58.7% | |
| 71.5% | 80% | 78% | 78% | 80% | 67% | 64% | 60% | 71% | 71% | 71% | 80% | 71% | 64% | 70% | 80% | 70% | 64% | 71% | 70% | 71% | |
| 56.1% | 61% | 63% | 63% | 68% | 80% | 46% | 53% | 43% | 53% | 58% | 53% | 46% | 53% | 53% | 69% | 61% | 51% | 48% | 48% | 52% | |
| 54.8% | 58% | 54% | 47% | 47% | 58% | 45% | 47% | 47% | 51% | 58% | 63% | 61% | 58% | 58% | 57% | 56% | 58% | 62% | 58% | 53% |