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 Kenya. 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 | 1st 84.5% | 8th 79.8% | 4th 82.5% | 4th 82.5% | 10th 77.3% | 19th 72.8% | 6th 81.5% | 16th 75.0% | 18th 73.3% | 9th 79.3% | 14th 76.0% | 3rd 83.3% | 11th 77.3% | 20th 72.3% | 11th 77.3% | 13th 76.8% | 21st 72.0% | 17th 74.8% | 7th 81.0% | 2nd 83.8% | 15th 75.3% | |
| 69.4% | 73% | 63% | 74% | 69% | 67% | 61% | 68% | 71% | 69% | 73% | 79% | 71% | 71% | 75% | 67% | 75% | 58% | 56% | 76% | 75% | 66% | |
| 84.6% | 98% | 83% | 86% | 88% | 91% | 88% | 85% | 89% | 81% | 74% | 86% | 94% | 88% | 77% | 90% | 81% | 81% | 83% | 85% | 82% | 66% | |
| 77.4% | 82% | 90% | 88% | 88% | 83% | 66% | 88% | 63% | 67% | 86% | 58% | 87% | 69% | 62% | 68% | 74% | 69% | 80% | 80% | 91% | 86% | |
| 80.6% | 85% | 83% | 82% | 85% | 68% | 76% | 85% | 77% | 76% | 84% | 81% | 81% | 81% | 75% | 84% | 77% | 80% | 80% | 83% | 87% | 83% |