# automl/SMAC3

SMAC3: A Versatile Bayesian Optimization Package for Hyperparameter Optimization

Repository: https://github.com/automl/SMAC3
Canonical: https://ross.abutalabs.com/products/smac3
Homepage: https://automl.github.io/SMAC3/latest/
Language: Python
License: NOASSERTION
License Family: other
Topics: bayesian-optimization, bayesian-optimisation, hyperparameter-optimization, hyperparameter-tuning, hyperparameter-search, configuration, algorithm-configuration, automl, automated-machine-learning, random-forest, gaussian-process
Last push: 2026-09-02T13:21:54+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 100, release rhythm 46, longevity 100
- inputs: {"age_days": 3668, "days_push": 0, "days_rel": 145, "gap_med": 222.0, "n_releases_24m": 3}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1244, forks 245 (observed 2026-09-03T02:15:09.372587+00:00)

## What it is
SMAC3 is a Python library for Bayesian Optimization used to tune hyperparameters of machine learning algorithms and configure arbitrary algorithms. It combines Bayesian optimization with an aggressive racing mechanism and supports multi-fidelity, multi-objective, and multi-threaded optimization via an ask-and-tell interface.

## Use cases
- tune hyperparameters of a machine learning model
- find the best configuration for my algorithm
- bayesian optimization with expensive objective functions
- multi-fidelity hyperparameter search
- multi-objective hyperparameter optimization
- automated machine learning hyperparameter tuning

## When to choose
- you need robust, actively maintained Bayesian optimization for hyperparameter or algorithm configuration
- you need multi-fidelity, multi-objective, or parallel (multi-threaded) optimization natively
- you want an ask-and-tell interface or resumable optimization runs
- you work in the AutoML ecosystem (Optuna, DeepCAVE integrations)

## When to avoid
- you need a command-line interface or runtime optimization, which were removed in v2.0
- you need distributed optimization across clusters - consider HyperSweeper or other tools instead
- you need simple random or grid search rather than model-based optimization

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, benchmarking, developer-tools
- domain: machine-learning, artificial-intelligence, data-science
- platform: python, windows, cross-platform
- tags: bayesian-optimization, hyperparameter-optimization, hyperparameter-tuning, automl, algorithm-configuration, random-forest, gaussian-process, multi-fidelity, multi-objective, ask-and-tell, algorithms, linux, macos

## Member repositories
- automl/SMAC3 (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-09-03T02:15:09.372587+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 2026-08-30T05:08:18.556661+00:00, confidence not recorded.
  - readme: https://github.com/automl/SMAC3 (fetched 2026-09-03T02:15:09.372587+00:00, sha 46490f1ccaca)
  - homepage: https://automl.github.io/SMAC3/latest/ (fetched 2026-08-29T12:20:06.987600+00:00, sha 9b9ed30fdd17)
- Data as of 2026-08-30T08:39:29.467469+00:00.
