# google/vizier

Python-based research interface for blackbox and hyperparameter optimization, based on the internal Google Vizier Service.

Repository: https://github.com/google/vizier
Canonical: https://ross.abutalabs.com/products/vizier
Homepage: https://oss-vizier.readthedocs.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: vizier, hyperparameter-optimization, tuning, tuning-parameters, blackbox-optimization, hyperparameter-tuning, bayesian-optimization, evolutionary-algorithms, distributed-systems, distributed-computing, grpc, open-source, google, algorithm, deep-learning, machine-learning, optimization
Last push: 2026-08-25T14:40:12+00:00

## Health v2 (maintenance only)
Score: 79/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 40, longevity 100
- inputs: {"age_days": 1659, "days_push": 8, "days_rel": 578, "gap_med": 5, "n_releases_24m": 6}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1671, forks 113 (observed 2026-08-28T04:05:20.151388+00:00)

## What it is
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
- artifact type: library
- maturity: active
- function: machine-learning, benchmarking, rpc, sdk
- domain: machine-learning, artificial-intelligence, deep-learning, microservices
- platform: python, cross-platform
- tags: hyperparameter-optimization, blackbox-optimization, bayesian-optimization, evolutionary-algorithms, hyperparameter-tuning, google-vizier, grpc-service, research, algorithms, docker

## Member repositories
- google/vizier (main) score 79

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:20.151388+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:42:41.024791+00:00, confidence not recorded.
  - readme: https://github.com/google/vizier (fetched 2026-08-28T04:05:20.151388+00:00, sha 3ae0241ced21)
- Data as of 2026-08-30T08:39:29.467469+00:00.