Google DeepMind·· 2026-05-06
AlphaEvolve: How our Gemini-powered coding agent is scaling impact across fields
AlphaEvolve: How our Gemini-powered coding agent is scaling impact across fields
AI summary
Google DeepMind published a year’s progress for AlphaEvolve. The Gemini-powered coding agent has expanded from open problems in mathematics and computer science into genomics, power grids, quantum physics and AI infrastructure.
Selection record
Threshold 60Official, first-handFirst 78Second 74
AdmittedSum of both 152 ≥ twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：Gemini驱动的编码智能体，AI算法发现与应用
- Why it was chosen
- A year of applications spanning genomics, power grids, quantum work and TPU design shows the cross-domain boundaries of algorithmic agents.
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: Google DeepMind · deepmind.google