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 South Africa. 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 | 3rd 79.8% | 9th 76.3% | 9th 76.3% | 8th 76.8% | 19th 70.5% | 21st 65.8% | 4th 78.0% | 17th 71.3% | 20th 69.5% | 14th 75.5% | 2nd 80.5% | 5th 77.3% | 16th 74.3% | 18th 70.8% | 7th 76.8% | 15th 75.3% | 11th 76.3% | 6th 77.0% | 13th 76.0% | 1st 81.7% | 11th 76.3% | |
| 69.7% | 71% | 77% | 68% | 60% | 68% | 54% | 65% | 58% | 71% | 71% | 77% | 78% | 77% | 77% | 78% | 82% | 62% | 66% | 73% | 72% | 58% | |
| 78.6% | 78% | 85% | 77% | 84% | 82% | 71% | 85% | 80% | 72% | 80% | 83% | 82% | 77% | 75% | 82% | 77% | 68% | 72% | 79% | 85% | 77% | |
| 68.5% | 74% | 66% | 74% | 78% | 72% | 59% | 75% | 64% | 55% | 71% | 70% | 66% | 59% | 57% | 55% | 64% | 75% | 75% | 76% | 78% | 75% | |
| 84.9% | 96% | 77% | 86% | 85% | 60% | 79% | 87% | 83% | 80% | 80% | 92% | 83% | 84% | 74% | 92% | 78% | 100% | 95% | 92% | 95% |