Google Cloud makes AlphaEvolve generally available

Google Cloud has made AlphaEvolve generally available on its Gemini Enterprise Agent Platform. Built on Gemini, the tool is a code optimisation and discovery agent.

The launch broadens access to a system previously limited to private preview and early access, where it was tested in logistics, semiconductors, genomics, financial services and computing research.

AlphaEvolve is designed for algorithmic problems where conventional software development struggles to search large numbers of possible implementations. Users provide a baseline algorithm, define how candidate programs should be scored, and run generated alternatives through their own evaluation environment before deploying the selected result into production.

The product has moved from an internal research effort into a tool used across Google's operations and by outside organisations. Within Google, it has been used to optimise silicon design for next-generation Tensor Processing Units, reduce write amplification in Google Spanner by 20%, and cut software storage footprints by nearly 9% through compiler changes.

Early users

Several companies reported gains in forecasting, routing, chip design and software performance.

BASF used the system to build a digital twin for its supply network after earlier attempts based on deterministic models failed. "We had several attempts to build a digital twin for our complex supply network using deterministic models, and all of them failed. By using AlphaEvolve, we can now not only map the complex network based on system data, but at the same time understand and copy the human decisions that drive our daily operations. This gives us a highly accurate and easy-to-maintain data-driven digital twin of the entire network," said Dr. Goetz Krabbe, Vice President for Global Supply Chain, BASF.

At Coolblue, the system was applied to a 28-day demand forecasting pipeline. "Coolblue data scientists used AlphaEvolve to directly optimize their 28-day demand forecasting pipeline, focusing on automated feature engineering, target preprocessing, and model selection. In just a few (200) iterations, AlphaEvolve improved our production forecast (by reducing WMAPE over the existing solution) by over 5%. These gains were achieved through improved feature engineering, an ensemble of different regression models, and better target preprocessing proposed and validated by AlphaEvolve. To ensure sufficient stock availability, it is crucial that the demand forecast is accurate for both the short term (the first 7 days) and the longer horizon (the full 28 days). AlphaEvolve achieved this by using an evaluation metric that combines both periods, along with a strict penalty for under forecasting. AlphaEvolve has proven its ability to significantly improve bulk purchasing decisions and help us maintain optimal stock levels for the weeks ahead," said Cas Ruger, Data Scientist, Coolblue.

FM Logistic reported a 10.4% improvement in warehouse routing. "Through our partnership with Google Cloud and the implementation of AlphaEvolve and Gemini, we further optimized our routing approach for fast-moving operations. The 10.4% improvement was achieved on top of an already highly optimized baseline, where further gains are typically hard to come by. This translates directly to faster fulfillment, improved working conditions for our teams, and reduced wear on our fleet," said Rodolphe Bey, Group CIO, FM Logistic.

Broader scope

In semiconductors, Infineon said early experiments showed potential across several stages of chip development. "Our initial experiments with AlphaEvolve have been very positive, demonstrating its potential to transform the chip design lifecycle. We see clear potential for it to contribute to multiple stages of development, including areas like surrogate modelling," said Michael Kollig, CIO, Infineon.

JetBrains used the tool for software performance work. "AlphaEvolve can change how we approach complex performance work. It turns optimizations that were once too time-consuming to explore into candidates we can test routinely. Engineers still own the benchmark, review, and release decision. The search space is what gets smaller," said Dmitrii Batkovich, Director of Engineering, JetBrains.

Kinaxis reported improvements in forecasting accuracy and runtime in early testing. "Kinaxis researchers have used AlphaEvolve to materially improve both the speed and quality of highly mature forecasting and optimization algorithms. In early testing, we achieved improvements of more than 22% in key forecasting accuracy metrics while reducing runtime by over 90% on benchmark datasets. As supply chains become increasingly complex and unpredictable, AlphaEvolve has the potential to help the world's largest organizations make faster, more informed decisions and adapt with greater confidence," said Gelu Ticala, Chief Technology Officer, Kinaxis.

Klarna applied the system to one of its largest machine learning training pipelines, where it doubled throughput while improving model quality under regulated operating conditions. Kuro Games reported performance gains in server-side workloads.

Research use

The tool is also being used in scientific computing and academic research. Oak Ridge National Labouratory deployed AlphaEvolve on Frontier, its exascale supercomputer, to optimise mixed-precision GPU kernels through repeated generation, compilation and validation of candidate programs on AMD GPUs.

"Our collaboration with Google's AlphaEvolve team gave us an early look at how evolutionary programming can be combined with leadership-class supercomputing. By running AlphaEvolve on Frontier, we explored a large number of optimization candidates in parallel, including novel implementation variants that helped us explore parts of the design space we might not have reached through manual optimization alone. This is an encouraging first step toward applying AI-assisted optimization to increasingly complex scientific software," said Oscar Hernandez Mendoza, PhD, Senior Computer Scientist, ORNL.

Old Dominion University used the system to search for Python programs that model biological ageing mortality rates, while PacBio said work with the tool produced higher sequencing accuracy for its instruments.

Other users named by Google included qBraid in quantum computing, Schrödinger in molecular simulation, Substrate in computational lithography, Pebble in GPU serving performance and WPP in campaign prediction models.

Pushmeet Kohli, Chief Scientist, Google Cloud & Vice President, Science at Google DeepMind, said: "AI is moving beyond acting as a productivity assistant that accelerates how we work to a discovery engine that expands what we can achieve. By autonomously navigating complex computational search spaces, tools like AlphaEvolve are helping researchers and engineers uncover breakthrough algorithms that augment traditional human intuition."