google/vizier

Health v2 · maintenance only

Adoption not part of the score

1671 stars · 113 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

Open Source Vizier is a Python-based service and library for black-box and hyperparameter optimization, based on Google's internal Vizier tuning service. It provides a client-server architecture supporting distributed multi-client optimization with algorithms like Bayesian optimization and evolutionary strategies.

Use cases

• tune hyperparameters for my deep learning model
• optimize a blackbox objective function
• run distributed hyperparameter search across multiple clients
• benchmark optimization algorithms
• find best parameters for a simulation
• bayesian optimization of expensive functions

When to choose

• you need scalable, distributed hyperparameter tuning backed by Google's production-proven service design
• you want a research platform for developing and benchmarking new optimization algorithms
• you need flexible search spaces with continuous, discrete, and categorical parameters

When to avoid

• you need a simple one-line AutoML solution with minimal setup
• your use case is gradient-based optimization of differentiable objectives
• you want a lightweight tuner without a client-server service architecture

Facets

library · maturity active

machine-learning benchmarking rpc sdk machine-learning artificial-intelligence deep-learning microservices python cross-platform hyperparameter-optimization blackbox-optimization bayesian-optimization evolutionary-algorithms hyperparameter-tuning google-vizier grpc-service research algorithms docker

Member repositories

For agents

markdown · JSON · MCP: product_card(name="google/vizier")