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 Korea South. 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 | 10th 81.5% | 7th 82.3% | 1st 87.5% | 17th 77.0% | 21st 74.3% | 11th 80.0% | 2nd 85.3% | 17th 77.0% | 20th 75.5% | 16th 78.5% | 6th 83.0% | 13th 79.3% | 4th 84.5% | 19th 76.3% | 8th 82.0% | 9th 81.8% | 5th 84.0% | 14th 79.0% | 12th 80.0% | 3rd 85.2% | 14th 79.0% | |
| 89.1% | 89% | 97% | 99% | 75% | 86% | 77% | 95% | 81% | 92% | 90% | 89% | 79% | 98% | 84% | 98% | 96% | 98% | 84% | 83% | 97% | 84% | |
| 87.8% | 82% | 84% | 98% | 78% | 78% | 91% | 96% | 91% | 78% | 76% | 96% | 88% | 91% | 93% | 91% | 87% | 92% | 96% | 83% | 93% | 81% | |
| 66.9% | 74% | 67% | 74% | 74% | 62% | 74% | 71% | 62% | 60% | 67% | 62% | 68% | 68% | 51% | 58% | 67% | 72% | 59% | 73% | 74% | 67% | |
| 78.7% | 81% | 81% | 79% | 81% | 71% | 78% | 79% | 74% | 72% | 81% | 85% | 82% | 81% | 77% | 81% | 77% | 74% | 77% | 81% | 77% | 84% |