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 Kazakhstan. 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 | 8th 78.8% | 20th 66.5% | 4th 80.8% | 2nd 82.3% | 6th 79.8% | 13th 76.0% | 7th 79.3% | 9th 78.0% | 11th 76.5% | 4th 80.8% | 16th 71.3% | 3rd 81.8% | 14th 75.3% | 12th 76.3% | 19th 67.8% | 15th 72.5% | 17th 71.0% | 18th 70.7% | 21st 39.2% | 10th 76.7% | 1st 83.3% | |
| 81.3% | 89% | 56% | 86% | 100% | 89% | 98% | 88% | 96% | 95% | 93% | 56% | 89% | 80% | 90% | 56% | 80% | 71% | 69% | 72% | 83% | 71% | |
| 71.7% | 73% | 70% | 82% | 81% | 83% | 68% | 75% | 73% | 75% | 65% | 75% | 75% | 70% | 77% | 73% | 76% | 83% | 70% | 10% | 76% | 75% | |
| 64.4% | 70% | 65% | 73% | 70% | 77% | 61% | 77% | 60% | 60% | 65% | 71% | 70% | 68% | 62% | 59% | 62% | 55% | 62% | 0% | 72% | 94% | |
| 80.6% | 83% | 75% | 82% | 78% | 70% | 77% | 77% | 83% | 76% | 100% | 83% | 93% | 83% | 76% | 83% | 72% | 75% | 82% | 75% | 76% | 93% |