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 Indonesia. 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 | 4th 85.5% | 3rd 85.8% | 11th 82.0% | 4th 85.5% | 17th 76.5% | 18th 76.3% | 9th 82.3% | 20th 74.3% | 19th 75.5% | 15th 79.3% | 1st 87.5% | 9th 82.3% | 6th 83.3% | 21st 69.5% | 16th 78.5% | 8th 82.7% | 12th 80.0% | 2nd 86.0% | 12th 80.0% | 7th 82.8% | 14th 79.5% | |
| 77.8% | 80% | 81% | 81% | 77% | 74% | 61% | 75% | 62% | 80% | 72% | 86% | 72% | 90% | 50% | 75% | 85% | 98% | 98% | 71% | 80% | 85% | |
| 87.9% | 89% | 96% | 80% | 87% | 87% | 83% | 88% | 91% | 94% | 86% | 98% | 96% | 85% | 91% | 87% | 91% | 79% | 82% | 95% | 84% | 77% | |
| 84.2% | 91% | 90% | 90% | 96% | 87% | 86% | 94% | 74% | 70% | 85% | 92% | 91% | 83% | 67% | 79% | 85% | 71% | 85% | 79% | 91% | 83% | |
| 72.9% | 82% | 76% | 77% | 82% | 58% | 75% | 72% | 70% | 58% | 74% | 74% | 70% | 75% | 70% | 73% | 70% | 72% | 79% | 75% | 76% | 73% |