ogb
Open Graph Benchmark
Decision gist · record as of 2026-08-14
Yes, if you are working on graph neural networks and need standardized benchmarks. The package is well-established with 2093 GitHub stars, has no known vulnerabilities, and provides genuine value through unified data loading and evaluation. However, maintenance is aging (last release 2023-04-07)—verify compatibility with your PyTorch and graph framework versions before relying on it for new research.AI-flagged interpretation of the facts on this page — verify before relying
Before you install
- Requires PyTorch>=1.6 and either DGL>=0.5.0 or torch-geometric>=2.0.2 installed separately; OGB itself does not bundle these.
- Low install friction with a pure-Python wheel.
- Maintenance is aging—last release was 2023-04-07, over 1225 days ago—but the repository remains active with 2093 stars and recent commits.
License · maintenance · safety
MIT (permissive) — MIT license is permissive, allowing commercial and private use with minimal restrictions. You must include a copy of the license and copyright notice.
last release 2023-04-07 (1225 days) · last repo commit 2025-05-06 · 2,093 stars
0 known vulnerabilities (OSV.dev, 2026-08-14) · 115,963 downloads/mo, #12,227 on PyPI
Alternatives
Verify before relying
pip install ogb
from ogb.graphproppred import PygGraphPropPredDataset
dataset = PygGraphPropPredDataset(name='ogbg-molhiv')
split_idx = dataset.get_idx_split()- Whether the aging maintenance status affects compatibility with recent PyTorch or graph framework versions.
- Current test coverage and whether benchmarks remain aligned with modern graph ML research standards.
- Actual monthly download volume and whether the 2093 GitHub stars reflect active research adoption.
What it is and what it does
OGB is a collection of standardized benchmark datasets and tools for evaluating graph machine learning methods. It provides easy-to-use data loaders compatible with popular graph deep learning frameworks, handling dataset downloading, standardized train/validation/test splits, and unified evaluation metrics. The package covers three core graph ML tasks—node prediction, link prediction, and graph prediction—across datasets of varying scales (from single-GPU to multi-GPU) and diverse domains including scientific networks, social networks, and heterogeneous knowledge graphs.
The package is designed to enable reliable comparison of different graph neural network methods by providing consistent dataset preparation and evaluation protocols. Researchers use OGB to benchmark their models against established baselines on real-world graph problems, avoiding the friction of manual dataset curation and metric implementation.
Use it for
- Benchmark a new graph neural network architecture against standardized datasets and compare results using OGB's unified evaluators.
- Download and prepare graph datasets with automatic splitting for training node, link, or graph classification models.
- Evaluate graph ML methods on datasets spanning different scales and domains without manual preprocessing.
- Reproduce published results on OGB benchmarks to validate research or baseline implementations.
- Access diverse real-world graphs from scientific, social, and knowledge-graph domains for algorithm development.
Worth the install?
AI-flagged interpretation of the facts on this page. Verify before relying on it.
Yes, if you are working on graph neural networks and need standardized benchmarks.
The package is well-established with 2093 GitHub stars, has no known vulnerabilities, and provides genuine value through unified data loading and evaluation. However, maintenance is aging (last release 2023-04-07)—verify compatibility with your PyTorch and graph framework versions before relying on it for new research.
Install
ogb on PyPI
Before you install
Low install friction with a pure-Python wheel. Maintenance is aging—last release was 2023-04-07, over 1225 days ago—but the repository remains active with 2093 stars and recent commits. Dependencies are stable and widely used.
Requires PyTorch>=1.6 and either DGL>=0.5.0 or torch-geometric>=2.0.2 installed separately; OGB itself does not bundle these.
License in practice
MIT license is permissive, allowing commercial and private use with minimal restrictions. You must include a copy of the license and copyright notice.
Quickstart
pip install ogb
from ogb.graphproppred import PygGraphPropPredDataset
dataset = PygGraphPropPredDataset(name='ogbg-molhiv')
split_idx = dataset.get_idx_split()Verify before relying
- Whether the aging maintenance status affects compatibility with recent PyTorch or graph framework versions.
- Current test coverage and whether benchmarks remain aligned with modern graph ML research standards.
- Actual monthly download volume and whether the 2093 GitHub stars reflect active research adoption.
Package facts
| License | MIT permissive |
| Python support | Not specified |
| Install friction | Low. Pure-Python wheel |
| Runtime dependencies | 8 packagestorchnumpytqdmscikit-learnpandassixurllib3outdated |
| Maintenance | Aging 1,225 days since the last release |
| Last repo commit | |
| First released | |
| Downloads | 115,963 / month, #12,227 on PyPI 30-day window, as of 2026-08-14 |
| Known vulnerabilities | None known OSV.dev, checked 2026-08-14 |
| Classifiers | Intended Audience :: Science/ResearchLicense :: OSI Approved :: MIT LicenseProgramming Language :: Python :: 3.6Programming Language :: Python :: 3.7Topic :: Scientific/Engineering :: Artificial Intelligence |
Evidence: ogb-1.3.6-py3-none-any.whl
Tags
Let your AI agent find packages like this
Example. Real query, live index.
You found this page by searching. An agent finds it by wishing: SkillFed indexes 14,416 PyPI packages by what they can do, searchable in plain language.
wish › “graph neural network benchmarks”
- ogbOGB provides standardized benchmark datasets, data loaders, and…
- jraphJraph provides data structures and utilities for building and working…
- torchvizGenerates visual diagrams of PyTorch neural network computation…
Give your agent the search over MCP, or paste the wish link into any chat.
More Artificial Intelligence packages
LiteLLM provides a unified Python interface to call 100+ LLM providers (OpenAI, Anthropic, Gemini, Bedrock, Azure, and others) using OpenAI-compatible API format, available as both a Python SDK and a self-hosted AI Gateway proxy server.
Install it if you need to work with multiple LLM providers or want to centralize LLM routing in your organization.
Client library and CLI tool for downloading, uploading, and managing models, datasets, and repositories on the Hugging Face Hub platform.
Install it if you work with Hugging Face Hub models or datasets.
LangChain provides a framework for building agents and LLM-powered applications by composing language models, tools, and memory through a unified API that abstracts over multiple model providers.
hf-xet provides chunk-based deduplication and efficient file transfer for the Hugging Face Hub, enabling faster uploads and downloads of large files with local disk caching.
Tokenizers converts raw text into token sequences for NLP models, with support for training custom vocabularies and using pre-built tokenizers (BPE, WordPiece) optimized for speed via Rust.
Transformers provides a unified framework for loading, fine-tuning, and running state-of-the-art pretrained models across text, vision, audio, video, and multimodal tasks using PyTorch, JAX, or TensorFlow.
Install it if you need to run or train any transformer-based model for NLP, vision, audio, or multimodal tasks.
See also pytdc · torch-geometric · clip-benchmark · mteb · recbole · rf100vl · objaverse · benchpots · datazets · arckit