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 Bangladesh. 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 | 9th 75.8% | 2nd 79.8% | 11th 74.8% | 20th 69.5% | 7th 76.8% | 16th 72.0% | 8th 76.0% | 2nd 79.8% | 18th 70.0% | 19th 69.8% | 10th 75.5% | 1st 80.0% | 4th 79.5% | 21st 66.3% | 5th 77.3% | 5th 77.3% | 14th 73.0% | 15th 72.8% | 17th 72.0% | 12th 74.5% | 13th 74.3% | |
| 56.6% | 65% | 65% | 60% | 53% | 51% | 47% | 58% | 51% | 53% | 50% | 64% | 64% | 65% | 44% | 63% | 67% | 57% | 44% | 63% | 54% | 50% | |
| 74.4% | 82% | 82% | 80% | 81% | 81% | 78% | 68% | 81% | 69% | 62% | 75% | 76% | 82% | 69% | 69% | 67% | 73% | 79% | 65% | 73% | 71% | |
| 80.4% | 75% | 79% | 80% | 75% | 87% | 78% | 85% | 89% | 79% | 78% | 74% | 91% | 78% | 74% | 85% | 79% | 83% | 76% | 78% | 78% | 87% | |
| 87.0% | 81% | 93% | 79% | 69% | 88% | 85% | 93% | 98% | 79% | 89% | 89% | 89% | 93% | 78% | 92% | 96% | 79% | 92% | 82% | 93% | 89% |