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 Judaism. 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 | 18th 75.8% | 10th 81.5% | 9th 82.0% | 14th 76.8% | 21st 73.8% | 17th 76.0% | 3rd 84.0% | 13th 80.3% | 19th 74.5% | 15th 76.5% | 2nd 84.3% | 16th 76.3% | 4th 84.0% | 20th 74.3% | 12th 81.0% | 8th 82.5% | 6th 82.8% | 11th 81.3% | 5th 83.3% | 1st 86.8% | 6th 82.8% | |
| 77.7% | 76% | 83% | 74% | 69% | 72% | 60% | 80% | 76% | 71% | 80% | 90% | 72% | 83% | 74% | 80% | 88% | 81% | 81% | 83% | 80% | ||
| 85.1% | 70% | 92% | 92% | 78% | 82% | 83% | 86% | 91% | 78% | 74% | 88% | 78% | 94% | 84% | 89% | 83% | 86% | 89% | 88% | 92% | 90% | |
| 68.8% | 68% | 67% | 73% | 73% | 72% | 70% | 75% | 66% | 59% | 66% | 70% | 69% | 69% | 63% | 65% | 66% | 66% | 66% | 74% | 75% | 72% | |
| 88.2% | 89% | 84% | 89% | 87% | 69% | 91% | 95% | 88% | 90% | 86% | 89% | 86% | 90% | 76% | 90% | 93% | 98% | 89% | 88% | 97% | 89% |