# ml-tooling/opyrator

🪄 Turns your machine learning code into microservices with web API, interactive GUI, and more.

Repository: https://github.com/ml-tooling/opyrator
Canonical: https://ross.abutalabs.com/products/opyrator
Homepage: https://opyrator-playground.mltooling.org
Language: Python
License: MIT
License Family: permissive
Topics: fastapi, streamlit, pydantic, python, microservices, serverless, faas, functions, python-functions, machine-learning, deployment, type-hints
Last push: 2026-08-25T02:35:35+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 8, longevity 100
- inputs: {"age_days": 1975, "days_push": 9, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 3132, forks 167 (observed 2026-08-28T04:07:45.556028+00:00)

## What it is
Opyrator is a Python library that turns plain Python functions into production-ready microservices with an auto-generated HTTP API (FastAPI) and interactive web UI (Streamlit). Services can be exported as portable executable files or Docker images and are built on OpenAPI, JSON Schema, and Python type hints.

## Use cases
- turn python functions into rest apis
- deploy machine learning models as microservices
- auto-generate a web ui for a python function
- package python code as a docker image for serving
- quickly prototype ml model demos with an api and gui
- share python functions as self-contained executable services

## When to choose
- you want to expose Python functions as HTTP APIs with minimal boilerplate
- you need a quick interactive UI for demos of ML models or functions
- you want portable, shareable service artifacts (files or Docker images) built on OpenAPI standards

## When to avoid
- you need a production-hardened, actively maintained framework - it is explicitly an alpha for experimental use
- you require fine-grained control over API design, auth, or complex multi-endpoint services
- your workload doesn't fit the single-function input/output model

## Facets
- artifact type: library
- maturity: experimental
- function: api-framework, web-framework, deployment, microservices, ui-components
- domain: machine-learning, web-development, developer-tools, apis
- platform: python, self-hosted
- tags: fastapi, streamlit, pydantic, faas, serverless, type-hints, openapi, function-as-a-service, docker, web-server

## Member repositories
- ml-tooling/opyrator (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:45.556028+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 2026-08-30T07:26:08.603485+00:00, confidence not recorded.
  - readme: https://github.com/ml-tooling/opyrator (fetched 2026-08-28T04:07:45.556028+00:00, sha df965fa6c34c)
  - registry_pypi: https://pypi.org/pypi/opyrator/json (fetched 2026-08-29T09:40:56.911555+00:00, sha 160bbb74abb8)
- Data as of 2026-08-30T08:39:29.467469+00:00.