Article complete

Google DeepMind Launches AlphaEvolve, New AI Coding Agent for Maths and Science

Google DeepMind is planning an early access programme for selected academic users and is also exploring ways to make AlphaEvolve more broadly available. 

A stylized light bulb icon in yellow with red, blue, and yellow backgrounds, symbolizing innovation and AI technology.

Key takeaways

  • Launch AlphaEvolve, a new AI coding agent for evolving complex algorithms in mathematics and science.
  • Integrate advanced language models with automated evaluators to optimize algorithmic solutions effectively.
  • Offer early access to selected academic users while exploring broader availability for general use.
  • Demonstrate transformative potential across fields like drug discovery, sustainability, and materials science.
  • Achieve significant efficiency gains, recovering 0.7% of global compute resources through optimized data center operations.

Takeaways drafted with AI from the article text.

Powered by Gemini Flash and Gemini Pro, AlphaEvolve pairs model-generated code with automated evaluators to verify, score, and evolve high-performing solutions.

“AlphaEvolve is an agent that can go beyond single function discovery to evolve entire codebases and develop much more complex algorithms,” stated Google DeepMind in its blog post. 

Google DeepMind is planning an early access programme for selected academic users and is also exploring ways to make AlphaEvolve more broadly available. Interested users can register their interest through a dedicated form.

The company believes AlphaEvolve could be transformative across multiple fields, including materials science, drug discovery, sustainability, and broader technological and business applications.

The system integrates prompt sampling, language model outputs, and program evaluation through an evolutionary algorithm framework.

Over the past year, AlphaEvolve has been used to improve data centre scheduling, hardware design, and AI training workflows across Google. One deployment optimised Borg, Google’s data centre orchestrator, recovering 0.7% of compute resources globally. “This solution, now in production for over a year, continuously recovers, on average, 0.7% of Google’s worldwide compute resources,” the company said.

AlphaEvolve also contributed to a Tensor Processing Unit (TPU) design. It suggested a Verilog-level change that removed redundant bits in a key arithmetic circuit.  Google said this proposal passed verification tests and was integrated into an upcoming TPU release.

In AI training, AlphaEvolve optimised matrix multiplication in the Gemini architecture, speeding up a core kernel by 23% and cutting training time by 1%. It also improved FlashAttention kernel performance by 32.5%, a domain typically untouched by human engineers due to compiler-level optimisation.

Beyond infrastructure, AlphaEvolve tackled algorithmic challenges in mathematics. It discovered a new method to multiply 4x4 complex-valued matrices using 48 scalar multiplications, improving on the 1969 Strassen algorithm. “This finding demonstrates a significant advance over our previous work, AlphaTensor,” the company said.

Applied to over 50 open problems across mathematics, AlphaEvolve rediscovered known solutions in 75% of cases and improved on 20%. One of its advances was in the kissing number problem, where it found a configuration of 593 spheres touching a unit sphere in 11 dimensions, establishing a new lower bound.

Published

Corrections: AIM corrects errors of fact and notes the change on the story. Spotted one? Tell the editors. Read our editorial standards.

More in this story

Google Deepmind · 5 stories
  1. Google DeepMind Introduces Semantica, An Adaptable Image-Conditioned Diffusion Model
  2. You are here: Google DeepMind Launches AlphaEvolve, New AI Coding Agent for Maths and Science
  3. Google DeepMind announces SignGemma: AI for Sign Language
  4. Google DeepMind Hires a Philosopher to Study Machine Consciousness & AGI
  5. Nobel-Winning AlphaFold Scientist John Jumper Leaves Google DeepMind for Anthropic
About the author

Tech Journalist

Siddharth is a media graduate who loves to explore tech through journalism and putting forward ideas worth pondering about in the era of artificial intelligence.

All stories
Belamy · Daily

The day’s AI news, before your first meeting.