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 Brazil. 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 | 8th 78.5% | 9th 78.3% | 1st 80.8% | 9th 78.3% | 2nd 79.8% | 18th 75.3% | 5th 79.3% | 16th 76.0% | 19th 74.5% | 20th 73.0% | 7th 78.7% | 9th 78.3% | 12th 77.0% | 12th 77.0% | 15th 76.3% | 4th 79.5% | 2nd 79.8% | 14th 76.5% | 21st 54.3% | 5th 79.3% | 17th 75.5% | |
| 85.9% | 89% | 83% | 94% | 91% | 86% | 65% | 91% | 78% | 88% | 83% | 96% | 85% | 85% | 83% | 92% | 98% | 84% | 76% | 89% | 84% | 84% | |
| 71.5% | 71% | 78% | 74% | 72% | 69% | 82% | 71% | 73% | 61% | 61% | 67% | 75% | 76% | 69% | 69% | 69% | 79% | 78% | 66% | 74% | 67% | |
| 65.9% | 65% | 76% | 67% | 65% | 77% | 66% | 72% | 65% | 54% | 57% | 76% | 67% | 58% | 59% | 62% | 60% | 69% | 70% | 62% | 71% | 65% | |
| 82.6% | 89% | 76% | 88% | 85% | 87% | 88% | 83% | 88% | 95% | 91% | 76% | 86% | 89% | 97% | 82% | 91% | 87% | 82% | 0% | 88% | 86% |