Scientific Idea Generation
96.67Novelty Win RateEvoScientist
Evaluation results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| EvoScientistEvaluator=Gemini-3-flash, System Category=Open-sourced Systems, Comparison Baseline=Virtual Scientist2026.03 | 96.67 | 3.33 | 0 | 93.33 | 6.67 | 0 | 90 | 6.67 | 3.33 | 96.67 | 3.33 | 0 | 93.34 | |
| EvoScientistEvaluator=Gemini-3-flash, System Category=Open-sourced Systems, Comparison Baseline=AI-Researcher2026.03 | 96.67 | 3.33 | 0 | 90 | 0 | 10 | 86.67 | 10 | 3.33 | 93.34 | 3.33 | 3.33 | 87.5 | |
| EvoScientistEvaluator=Gemini-3-flash, System Category=Commercial Systems, Comparison Baseline=Hypogenic2026.03 | 93.33 | 6.67 | 0 | 83.34 | 3.33 | 13.33 | 70 | 23.33 | 6.67 | 96.67 | 3.33 | 0 | 80.83 | |
| EvoScientistEvaluator=Gemini-3-flash, System Category=Commercial Systems, Comparison Baseline=Novix2026.03 | 90 | 6.67 | 3.33 | 53.33 | 10 | 36.67 | 46.67 | 36.66 | 16.67 | 70.67 | 10 | 20 | 46 | |
| EvoScientistEvaluator=Gemini-3-flash, System Category=Commercial Systems, Comparison Baseline=K-Dense2026.03 | 86.67 | 3.33 | 10 | 56.67 | 13.33 | 30 | 43.33 | 36.67 | 20 | 76.67 | 13.33 | 10 | 54.5 | |
| EvoScientistEvaluator=Gemini-3-flash, System Category=Open-sourced Systems, Comparison Baseline=InternAgent2026.03 | 73.33 | 16.67 | 10 | 93.33 | 0 | 6.67 | 86.67 | 13.33 | 0 | 96.67 | 3.33 | 0 | 83.33 | |
| EvoScientistEvaluator=Gemini-3-flash, System Category=Open-sourced Systems, Comparison Baseline=AI Scientist-v22026.03 | 63.33 | 16.67 | 20 | 53.33 | 6.67 | 40 | 36.67 | 50 | 13.33 | 56.67 | 23.33 | 20 | 29.17 |