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 Turkey. 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 | 12th 78.5% | 4th 80.8% | 14th 77.3% | 6th 80.0% | 16th 76.5% | 21st 69.3% | 7th 79.8% | 17th 75.0% | 17th 75.0% | 17th 75.0% | 9th 79.0% | 15th 77.3% | 3rd 81.8% | 1st 83.8% | 2nd 83.3% | 5th 80.5% | 12th 78.5% | 20th 73.5% | 8th 79.5% | 9th 79.0% | 9th 79.0% | |
| 63.5% | 58% | 72% | 59% | 52% | 59% | 45% | 63% | 52% | 68% | 50% | 68% | 57% | 72% | 80% | 72% | 77% | 68% | 62% | 71% | 64% | 65% | |
| 82.0% | 80% | 94% | 76% | 78% | 80% | 77% | 79% | 78% | 80% | 78% | 90% | 78% | 78% | 100% | 94% | 80% | 78% | 68% | 80% | 94% | 82% | |
| 87.1% | 97% | 75% | 93% | 95% | 96% | 75% | 98% | 91% | 81% | 95% | 79% | 95% | 88% | 76% | 88% | 83% | 88% | 80% | 88% | 79% | 90% | |
| 80.1% | 79% | 82% | 81% | 95% | 71% | 80% | 79% | 79% | 71% | 77% | 79% | 79% | 89% | 79% | 79% | 82% | 80% | 84% | 79% | 79% | 79% |