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 Christianity. 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 | 5th 86.3% | 13th 83.3% | 2nd 88.8% | 7th 85.5% | 18th 82.5% | 9th 84.3% | 6th 86.0% | 20th 82.0% | 21st 82.0% | 10th 84.0% | 1st 91.5% | 12th 83.5% | 14th 83.2% | 11th 83.7% | 4th 86.8% | 19th 82.3% | 15th 83.0% | 8th 84.7% | 17th 82.5% | 2nd 88.8% | 16th 82.8% | |
| 77.8% | 78% | 82% | 76% | 66% | 79% | 68% | 77% | 76% | 77% | 77% | 85% | 73% | 81% | 83% | 86% | 85% | 75% | 74% | 76% | 83% | 76% | |
| 85.2% | 88% | 83% | 90% | 87% | 82% | 87% | 87% | 90% | 85% | 78% | 90% | 83% | 86% | 85% | 89% | 82% | 86% | 85% | 84% | 83% | 79% | |
| 83.1% | 85% | 73% | 92% | 93% | 92% | 89% | 89% | 70% | 76% | 88% | 92% | 86% | 76% | 71% | 75% | 73% | 82% | 88% | 80% | 93% | 82% | |
| 92.5% | 94% | 95% | 97% | 96% | 77% | 93% | 91% | 92% | 90% | 93% | 99% | 92% | 90% | 96% | 97% | 89% | 89% | 92% | 90% | 96% | 94% |