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 Israel. 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 | 17th 76.5% | 13th 78.8% | 10th 79.8% | 16th 77.8% | 13th 78.8% | 9th 79.8% | 7th 80.8% | 12th 79.5% | 18th 75.5% | 19th 75.3% | 6th 81.5% | 13th 78.8% | 1st 85.5% | 20th 73.8% | 11th 79.5% | 5th 81.5% | 2nd 84.3% | 4th 82.0% | 21st 53.3% | 3rd 84.0% | 8th 80.0% | |
| 79.3% | 73% | 85% | 76% | 70% | 72% | 68% | 80% | 70% | 77% | 76% | 90% | 78% | 87% | 76% | 84% | 90% | 80% | 85% | 86% | 83% | ||
| 79.0% | 68% | 83% | 88% | 80% | 80% | 94% | 81% | 88% | 75% | 71% | 83% | 75% | 94% | 83% | 86% | 78% | 94% | 87% | 0% | 89% | 83% | |
| 66.8% | 72% | 65% | 70% | 72% | 71% | 68% | 71% | 61% | 56% | 62% | 77% | 70% | 70% | 60% | 63% | 60% | 66% | 65% | 69% | 69% | 66% | |
| 89.7% | 93% | 82% | 85% | 89% | 92% | 89% | 91% | 99% | 94% | 92% | 76% | 92% | 91% | 76% | 85% | 98% | 97% | 91% | 91% | 92% | 88% |