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 Islam. 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 | 6th 87.0% | 2nd 88.8% | 8th 86.5% | 10th 86.5% | 18th 80.8% | 21st 77.3% | 1st 89.8% | 19th 80.5% | 20th 79.5% | 17th 81.3% | 3rd 88.3% | 13th 84.3% | 11th 85.5% | 14th 83.5% | 5th 87.3% | 15th 83.3% | 11th 85.5% | 16th 83.0% | 3rd 88.3% | 8th 86.5% | 7th 86.8% | |
| 75.1% | 73% | 78% | 72% | 64% | 71% | 59% | 82% | 70% | 78% | 66% | 84% | 68% | 82% | 79% | 83% | 81% | 79% | 74% | 83% | 72% | 79% | |
| 88.0% | 94% | 90% | 86% | 90% | 88% | 78% | 90% | 89% | 86% | 79% | 95% | 92% | 88% | 86% | 93% | 93% | 86% | 86% | 91% | 86% | 82% | |
| 88.2% | 88% | 95% | 95% | 94% | 95% | 83% | 97% | 78% | 80% | 92% | 83% | 91% | 84% | 78% | 84% | 82% | 92% | 81% | 92% | 95% | 94% | |
| 87.7% | 93% | 92% | 93% | 98% | 69% | 89% | 90% | 85% | 74% | 88% | 91% | 86% | 88% | 91% | 89% | 77% | 85% | 91% | 87% | 93% | 92% |