---
id: objectgraph
version: "1.0.5"
license: MIT
license_treatment: permissive
maintenance: aging
---

# objectgraph — A basic graph library

License: permissive · Maintenance: aging · Downloads: 129.4K/mo

## What it is and what it does

Objectgraph is a graph library for Python 3 that lets you represent directed graphs where nodes are arbitrary objects (identified by a hashable 'identifier' attribute) and edges can carry arbitrary attributes. The library uses a non-standard edge model: multiple edges between the same pair of nodes are allowed, but edges with identical attributes are automatically collapsed into a single edge, reducing redundancy.

The package is a complete rewrite of the ObjectGraph class from altgraph, built from scratch for Python 3 with full test coverage. It depends only on typing_extensions and supports Python 3.10 through 3.15, including free-threading support in Python 3. The codebase is marked Production/Stable and has been actively maintained since its initial release in 2019, though recent activity has slowed.

Use it for:

- Build dependency graphs or call graphs where you need to track relationships between arbitrary Python objects.
- Model network topologies or system architectures where nodes represent entities and edges represent connections with metadata.
- Analyze object relationships in memory profiling or debugging tools that need to represent object references.
- Implement graph-based algorithms on custom domain objects without writing graph infrastructure from scratch.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Objectgraph provides a Python class for representing directed graphs with arbitrary objects as nodes and multi-edge support, where edges with identical attributes collapse into a single edge.

Yes, if you need a lightweight, permissively licensed graph library for Python 3.10+. The single dependency, Production/Stable status, and zero known vulnerabilities make it low-risk. The aging maintenance status (no release in 266 days) is a minor concern for a stable library but worth monitoring if you plan long-term use. Suitable for small to medium graph use cases; verify performance characteristics for larger workloads.

## Install

pip install objectgraph
uv add objectgraph
poetry add objectgraph

## Installing objectgraph

Before you install: Low install friction; single lightweight dependency on typing_extensions. Maintenance status is aging—last commit was 2026-01-02 and no release in 266 days—but the codebase is marked Production/Stable and supports current Python versions through 3.15.

License in practice: MIT license (permissive) places no restrictions on use, modification, or distribution in commercial or private projects.

Quickstart:

pip install objectgraph

from objectgraph import ObjectGraph

graph = ObjectGraph()
# Add nodes and edges via the graph API

Requires Python 3.10 or later.

Verify before relying:

- Whether the graph implementation is optimized for large-scale use or primarily suited to small to medium graphs.
- Performance characteristics compared to other graph libraries or whether benchmarks exist.
- Whether the 'identifier' attribute requirement for nodes imposes practical constraints on object types.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: aging
- Downloads: 129.4K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

graph library python, directed graph nodes edges, object graph data structure, graph representation library, network graph python, graph with arbitrary objects, multi-edge graph support, graph-algorithms, data-structures

[View on SkillFed](https://skillfed.io/packages/objectgraph) · [View on PyPI](https://pypi.org/project/objectgraph/)