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 | 4th 78.3% | 17th 71.5% | 10th 76.3% | 9th 76.8% | 18th 70.5% | 21st 65.8% | 2nd 79.8% | 19th 69.8% | 20th 68.8% | 14th 73.3% | 1st 80.0% | 6th 77.8% | 11th 75.0% | 15th 73.0% | 8th 77.0% | 12th 74.0% | 16th 72.0% | 12th 74.0% | 3rd 79.5% | 5th 78.0% | 7th 77.3% | |
| 67.1% | 71% | 60% | 68% | 60% | 67% | 54% | 71% | 65% | 67% | 61% | 62% | 69% | 72% | 77% | 78% | 82% | 66% | 62% | 72% | 66% | 59% | |
| 80.3% | 84% | 85% | 78% | 84% | 77% | 71% | 83% | 80% | 77% | 77% | 83% | 85% | 82% | 84% | 84% | 77% | 72% | 73% | 84% | 85% | 81% | |
| 67.6% | 74% | 66% | 74% | 78% | 71% | 59% | 77% | 51% | 51% | 71% | 80% | 74% | 62% | 57% | 54% | 59% | 66% | 77% | 78% | 69% | 72% | |
| 83.7% | 84% | 75% | 85% | 85% | 67% | 79% | 88% | 83% | 80% | 84% | 95% | 83% | 84% | 74% | 92% | 78% | 84% | 84% | 84% | 92% | 97% |