aioprometheus

What it is and what it does

aioprometheus is a Prometheus client library designed for Python asyncio applications. It lets you instrument your code by creating metrics (counters, gauges, histograms, etc.) and then expose them to Prometheus or compatible monitoring systems. The library includes a default collector registry and supports both text export and push-gateway modes.

For web frameworks, aioprometheus provides ASGI middleware that automatically tracks request counts, response codes, and exceptions in FastAPI, Starlette, and Quart applications. For standalone asyncio services, you can use the HTTP Service to expose metrics on a dedicated endpoint. The core library has minimal dependencies; framework-specific features require optional extras (aiohttp, starlette, quart).

Use it for:

• Add Prometheus metrics to a FastAPI or Starlette application via ASGI middleware to track HTTP request volume and response codes
• Instrument a long-running asyncio service (e.g., worker, daemon) with custom counters and expose them via an HTTP metrics endpoint
• Push metrics from a distributed asyncio process to a Prometheus Push Gateway for scrape-less collection
• Track application events and performance in asyncio-based microservices without blocking the event loop
• Monitor custom business logic (e.g., job counts, queue depth) alongside standard HTTP metrics in async web apps

Worth the install?

AI-flagged interpretation of the facts on this page — verify before relying

A Prometheus client library for asyncio applications that collects and exposes metrics via text export or gateway push, with optional ASGI middleware for FastAPI, Starlette, and Quart.

Yes, with conditions. aioprometheus is stable and production-ready for asyncio applications that need Prometheus integration. Install it if you are building or maintaining an asyncio-based service (FastAPI, Starlette, Quart, or standalone) and need metrics collection. The dormant maintenance status is not a blocker for stable use, but verify compatibility with your Prometheus server version and Python 3.8+ support meets your needs. No known security vulnerabilities.

Install

aioprometheus on PyPI

pip install aioprometheus

uv add aioprometheus

poetry add aioprometheus

Installing aioprometheus

Before you install

Low install friction with only two runtime dependencies (orjson and quantile-python). Maintenance is dormant—last release was 961 days ago—but the project remains archived=false with a recent commit (2024-08-21) and is marked Production/Stable. Suitable for stable, long-running deployments where active development is not expected.

License in practice

MIT license (permissive) means you can use, modify, and distribute aioprometheus with minimal restrictions, including in commercial applications. No copyleft obligations.

Quickstart

pip install aioprometheus

from aioprometheus import Counter

events_counter = Counter("events_counter", "Total number of events.")
events_counter.inc({"kind": "event A"})

Requires Python >=3.8.0. ASGI middleware extras (starlette, quart, aiohttp) are optional but needed for framework-specific features.

Verify before relying

• Whether quantile-python and orjson versions are pinned or have known compatibility constraints
• Performance characteristics under high-volume metric collection in production asyncio workloads
• Whether dormant maintenance status affects compatibility with recent Prometheus server versions

Package facts

Evidence: aioprometheus-23.12.0-py3-none-any.whl

Keywords: prometheus, monitoring, metrics

Tags

More Python Modules packages