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 | 7th 84.2% | 8th 83.5% | 2nd 87.5% | 18th 76.3% | 21st 72.5% | 12th 80.0% | 3rd 87.3% | 15th 77.5% | 20th 74.5% | 17th 76.5% | 4th 87.0% | 13th 79.3% | 1st 87.8% | 19th 76.0% | 16th 76.8% | 10th 81.8% | 14th 78.3% | 9th 82.8% | 11th 80.8% | 5th 85.8% | 6th 85.5% | |
| 90.1% | 94% | 97% | 99% | 75% | 81% | 77% | 98% | 81% | 92% | 92% | 97% | 79% | 98% | 84% | 94% | 96% | 90% | 98% | 89% | 97% | 84% | |
| 87.5% | 88% | 89% | 98% | 78% | 72% | 91% | 96% | 93% | 78% | 76% | 95% | 88% | 98% | 93% | 81% | 87% | 83% | 89% | 92% | 91% | 81% | |
| 67.8% | 74% | 67% | 74% | 74% | 67% | 74% | 74% | 62% | 60% | 61% | 71% | 68% | 74% | 50% | 58% | 67% | 63% | 67% | 68% | 74% | 77% | |
| 78.7% | 81% | 81% | 79% | 78% | 70% | 78% | 81% | 74% | 68% | 77% | 85% | 82% | 81% | 77% | 74% | 77% | 77% | 77% | 74% | 81% | 100% |