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 More concerned than excited. 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 87.3% | 13th 75.8% | 7th 78.8% | 14th 75.0% | 16th 74.0% | 9th 78.3% | 20th 67.3% | 2nd 83.5% | 8th 78.5% | 5th 80.0% | 19th 71.0% | 11th 77.5% | 12th 76.0% | 21st 59.3% | 15th 74.3% | 18th 73.3% | 16th 74.0% | 3rd 82.2% | 6th 79.0% | 10th 77.8% | 4th 80.8% | |
| 59.0% | 100% | 48% | 54% | 57% | 76% | 51% | 55% | 100% | 56% | 46% | 26% | 61% | 55% | 59% | 57% | 55% | 51% | 58% | 67% | 49% | 59% | |
| 87.6% | 93% | 89% | 83% | 82% | 82% | 93% | 88% | 90% | 95% | 97% | 80% | 95% | 87% | 67% | 80% | 93% | 79% | 97% | 82% | 94% | 93% | |
| 69.2% | 73% | 75% | 81% | 76% | 58% | 75% | 58% | 55% | 72% | 77% | 89% | 66% | 63% | 35% | 63% | 54% | 70% | 80% | 76% | 76% | 82% | |
| 89.5% | 83% | 91% | 97% | 85% | 80% | 94% | 68% | 89% | 91% | 100% | 89% | 88% | 99% | 76% | 97% | 91% | 96% | 94% | 91% | 92% | 89% |