Oríon

Spec sheet

Purpose

Oríon is an asynchronous framework for black-box function optimization, including hyperparameter optimization for machine learning.Read 11 Oct 2026 · orion.readthedocs.io

Intended users

The documentation describes Oríon as useful for machine learning researchers and engineers, while saying it is not itself a machine learning framework, pipeline, or automated machine learning product.Read 11 Oct 2026 · orion.readthedocs.io

Platforms

The installation guide says Oríon supports most Linux distributions and Mac OS X, and Windows 10 through Windows Subsystem for Linux.Read 11 Oct 2026 · orion.readthedocs.io

Support

The project directs users to report questions, issues, bugs, or feature suggestions through its GitHub issue tracker.Read 11 Oct 2026 · orion.readthedocs.io

License

The project documentation says Oríon is licensed under the BSD license.Read 4 Oct 2026 · orion.readthedocs.io

Algorithms

The installation guide lists Random Search, Hyperband, TPE, and ASHA as included algorithms, with more available through plugins.Read 4 Oct 2026 · orion.readthedocs.io

Database backends

The documented built-in backends are EphemeralDB, file-based PickledDB, and MongoDB; MongoDB is recommended for large-scale parallel optimization above 50 workers.Read 11 Oct 2026 · orion.readthedocs.io

Experiment tracking

Oríon stores trials in a configured database and offers experiment versioning to track trials and their parameters for reproducibility.Read 11 Oct 2026 · orion.readthedocs.io

Installation

The installation guide directs users to install Oríon from PyPI with pip install orion.Read 11 Oct 2026 · orion.readthedocs.io

Integration

A script can report its optimization objective through Oríon’s Python client API.Read 4 Oct 2026 · orion.readthedocs.io

Optimization algorithms

The documentation lists Random Search, Hyperband, TPE, and ASHA as included algorithms, and supports adding algorithms through plugins.Read 11 Oct 2026 · orion.readthedocs.io

Parallel execution

Oríon is natively asynchronous and the documentation says it can run in parallel from a laptop to a computing farm with thousands of processors.Read 11 Oct 2026 · orion.readthedocs.io

Scaling limit

The documentation says PickledDB is generally not suited to parallel optimization with more than 50 workers, though it describes this as a rule of thumb.Read 11 Oct 2026 · orion.readthedocs.io

Search spaces

Oríon supports search spaces defined by priors, including almost all SciPy distributions, configured through the command line or a configuration file.Read 11 Oct 2026 · orion.readthedocs.io

Storage

The documented database backends are in-memory EphemeralDB, file-based PickledDB, and MongoDB.Read 4 Oct 2026 · orion.readthedocs.io

Storage limit

The documentation says PickledDB is generally not suited to parallel optimization with more than 50 workers; MongoDB is recommended above that scale.Read 4 Oct 2026 · orion.readthedocs.io

Visualization

The documentation includes experiment visualizations such as regret curves, parallel coordinates, local parameter importance, and partial dependencies.Read 4 Oct 2026 · orion.readthedocs.io

Workflow

It can tune command-line scripts with minimal changes to a researcher’s existing workflow.Read 4 Oct 2026 · orion.readthedocs.io

Workflow integration

The documentation says users can integrate Oríon into an existing workflow by calling report_objective() at the end of their script, or use its Python API.Read 11 Oct 2026 · orion.readthedocs.io

Where Specifiction places it

• ML Experiment Tracking Software#25 of 397.1