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 | 1st 83.5% | 3rd 79.5% | 4th 78.8% | 19th 69.5% | 15th 73.8% | 18th 69.8% | 8th 77.5% | 6th 78.3% | 13th 75.0% | 17th 70.5% | 9th 77.3% | 2nd 80.8% | 5th 78.5% | 20th 69.0% | 6th 78.3% | 10th 77.0% | 14th 74.8% | 16th 71.0% | 21st 33.5% | 12th 75.8% | 11th 76.0% | |
| 59.4% | 65% | 63% | 60% | 53% | 57% | 47% | 64% | 51% | 59% | 50% | 64% | 64% | 65% | 55% | 70% | 62% | 61% | 57% | 61% | 51% | 69% | |
| 74.1% | 97% | 82% | 81% | 81% | 72% | 78% | 66% | 81% | 65% | 65% | 69% | 76% | 81% | 69% | 69% | 61% | 66% | 70% | 73% | 81% | 74% | |
| 77.8% | 91% | 80% | 80% | 75% | 87% | 69% | 87% | 83% | 84% | 78% | 80% | 91% | 75% | 74% | 85% | 89% | 91% | 84% | 0% | 78% | 72% | |
| 83.5% | 81% | 93% | 94% | 69% | 79% | 85% | 93% | 98% | 92% | 89% | 96% | 92% | 93% | 78% | 89% | 96% | 81% | 73% | 0% | 93% | 89% |