Koha vs Omeka S vs Invenio: Self-Hosted Library & Digital Collection Systems 2026

Libraries, archives, and cultural heritage institutions need robust software to manage catalogs, circulate materials, digitize collections, and provide public access to their holdings. Commercial integrated library systems (ILS) and digital repository platforms cost tens of thousands of dollars annually and lock institutions into proprietary data formats. Self-hosted open-source library platforms give institutions complete control over their collections, metadata standards, and public interfaces.

This guide compares three leading open-source platforms for different library use cases: Koha for traditional integrated library management, Omeka S for digital collections and exhibits, and Invenio for research-grade digital repositories.

Comparison at a Glance

Koha — Integrated Library System

Koha is the world’s most widely used open-source integrated library system. Originally developed in 1999 by Katipo Communications for the Horowhenua Library Trust in New Zealand, Koha is now maintained by a global community of librarians and developers. It handles the full lifecycle of library operations — from cataloging and circulation to public access and reporting.

Core features:

• Cataloging — Full MARC21 and UNIMARC support, Z39.50/SRU searching for record import, authority control, and classification with DDC, LCC, or custom schemes
• Circulation management — Check-in/check-out, holds/reserves, fines and fees, patron accounts, and circulation rules based on patron category and item type
• OPAC (Online Public Access Catalog) — Searchable public catalog with faceted search, RSS feeds, social sharing, user reviews, and tag clouds
• Acquisitions — Budget management, vendor tracking, purchase orders, and invoice processing
• Serials management — Subscription tracking, issue prediction, and receipt management for periodicals and journals
• Patron management — Member registration, category-based permissions, borrowing limits, and messaging
• Reports and statistics — Custom SQL reports, pre-built statistical reports, and data export for external analysis
• Self-checkout — Web-based self-checkout kiosk interface for patron self-service
• Barcode and RFID support — Integration with barcode scanners and RFID systems for automated circulation
• Multi-branch support — Consortium mode with independent or shared catalogs across multiple library branches

Strengths: Koha is the most complete open-source ILS available, handling everything from cataloging to circulation to acquisitions. The MARC21 cataloging module meets professional library standards. The multi-branch consortium mode makes it ideal for library networks. A large global community provides plugins, translations, and support. The OPAC is fully customizable with templates and CSS.

Limitations: Koha is a full ILS — it is overkill for institutions that only need a digital collection showcase. The MARC cataloging workflow requires trained catalogers. The Perl-based codebase can be challenging for developers unfamiliar with the ecosystem.

Docker Deployment for Koha

The Koha community provides Docker images for deployment:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32

version: '3.8'

services:
  koha:
    image: koha/koha-community:latest
    ports:
      - "8080:80"
      - "8081:8081"
    environment:
      - KOHA_DOMAIN=localhost
      - DB_HOST=mysql
      - DB_PASSWORD=koha_secret
    volumes:
      - koha_data:/var/lib/koha
    depends_on:
      - mysql
    restart: unless-stopped

  mysql:
    image: mysql:8.0
    environment:
      - MYSQL_ROOT_PASSWORD=root_secret
      - MYSQL_DATABASE=koha
      - MYSQL_USER=koha
      - MYSQL_PASSWORD=koha_secret
    volumes:
      - mysql_data:/var/lib/mysql
    restart: unless-stopped

volumes:
  koha_data:
  mysql_data:

After deployment, Koha’s web installer at http://localhost:8080 guides you through library configuration: setting up the MARC framework, defining patron categories, creating item types, and configuring circulation rules. The staff client runs on port 8081 and the public OPAC on port 8080.

Omeka S — Digital Collections & Exhibits

Omeka S is a web publishing platform for sharing digital collections and creating online exhibits. Developed by the Roy Rosenzweig Center for History and New Media at George Mason University, Omeka S is designed for cultural heritage institutions — museums, galleries, archives, and libraries — that need to display digitized materials with rich metadata and curated narratives.

Core features:

• Item management — Upload and catalog digital objects (images, documents, audio, video) with customizable metadata fields using Dublin Core and extended schemas
• Media support — Multiple media types per item, including images, PDFs, audio files, video, and 3D objects (via IIIF and embedded viewers)
• Site builder — Drag-and-drop page builder for creating public-facing exhibit websites with custom themes and layouts
• Multi-site architecture — Create multiple independent public sites from a single Omeka S installation, each with its own theme, pages, and content
• Linked data — Built-in support for linked open data, enabling connections between items and external vocabulary services (VIAF, Getty AAT, Library of Congress)
• Module ecosystem — Extensible architecture with 50+ modules for IIIF support, geolocation, bulk metadata editing, PDF generation, and more
• User roles — Global admin, site admin, creator, reviewer, and reader roles for collaborative content management
• OAI-PMH harvesting — Export metadata for discovery through aggregators and union catalogs
• REST API — Full API for programmatic access to items, media, sites, and metadata

Strengths: Omeka S excels at creating visually compelling digital exhibits. The multi-site architecture lets institutions host multiple curated collections from a single installation. The module ecosystem is rich and actively maintained. Linked data support makes collections discoverable across institutions. The platform is used by major museums, universities, and archives worldwide.

Limitations: Omeka S is not an ILS — it does not handle circulation, patron management, or MARC cataloging. It is best used alongside a traditional ILS (like Koha) for libraries that need both catalog management and digital exhibition capabilities.

Docker Deployment for Omeka S

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37

version: '3.8'

services:
  omeka:
    image: omeka/omeka-s:latest
    ports:
      - "8082:80"
    environment:
      - OMEKA_S_DB_HOST=mysql
      - OMEKA_S_DB_NAME=omeka
      - OMEKA_S_DB_USER=omeka
      - OMEKA_S_DB_PASSWORD=omeka_secret
      - OMEKA_S_INSTALL=true
    volumes:
      - omeka_files:/var/www/html/files
      - omeka_modules:/var/www/html/modules
      - omeka_themes:/var/www/html/themes
    depends_on:
      - mysql
    restart: unless-stopped

  mysql:
    image: mysql:8.0
    environment:
      - MYSQL_ROOT_PASSWORD=root_secret
      - MYSQL_DATABASE=omeka
      - MYSQL_USER=omeka
      - MYSQL_PASSWORD=omeka_secret
    volumes:
      - mysql_data:/var/lib/mysql
    restart: unless-stopped

volumes:
  omeka_files:
  omeka_modules:
  omeka_themes:
  mysql_data:

Omeka S installation is web-based. After starting the containers, visit http://localhost:8082 to create the first administrator account and configure the database. The platform ships with several default themes and modules that can be extended via the admin dashboard.

Invenio — Research-Grade Digital Repository

Invenio is an open-source framework for building large-scale digital repositories, developed by CERN for managing scientific publications, research data, and institutional archives. It powers CERN’s Document Server (over 2 million records) and is used by universities and research organizations worldwide through the InvenioRDM distribution.

Core features:

• Flexible metadata schemas — Support for MARC21, Dublin Core, DataCite, and custom JSON schemas. Schema can be customized per community or collection
• File management — Large file storage with versioning, checksums, and preservation metadata. Support for millions of files with efficient storage backends
• OAI-PMH server — Native OAI-PMH harvesting interface for metadata sharing with aggregators like BASE, CORE, and Google Scholar
• REST API — Comprehensive RESTful API for records, files, communities, and workflows. GraphQL support for flexible querying
• Communities and collections — Multi-tenant architecture with communities (research groups, departments, projects) managing their own collections within a shared repository
• Access control — Fine-grained access management with public, embargoed, and restricted records. Role-based permissions for depositors, reviewers, and administrators
• DOI minting — Integration with DataCite and CrossRef for automatic DOI assignment and registration
• Statistics — Download and view statistics at the record, file, and community level. Integration with Matomo and Google Analytics
• Search and discovery — Elasticsearch-powered search with faceting, autocomplete, and advanced query syntax. Customizable discovery interface
• Workflow management — Submission, review, and publication workflows with configurable approval steps

Strengths: Invenio is designed for scale — it handles millions of records and petabytes of data. The flexible metadata system supports any schema, making it suitable for research data, institutional repositories, and digital archives. The CERN-backed development ensures long-term stability and professional support options. The InvenioRDM distribution provides a ready-to-use research data management platform.

Limitations: Invenio has a steeper learning curve than Koha or Omeka S. It requires Elasticsearch and PostgreSQL, adding infrastructure complexity. The default interface is functional but requires customization to match institutional branding. It is not designed for circulation or patron management.

Docker Deployment for Invenio

Invenio provides an official Docker Compose setup for development and production:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59

version: '3.8'

services:
  frontend:
    image: inveniosoftware/invenio-app-frontend:latest
    ports:
      - "8083:5000"
    environment:
      - INVENIO_FRONTEND_API_URL=http://api:5000/api
    depends_on:
      - api
    restart: unless-stopped

  api:
    image: inveniosoftware/invenio-app-api:latest
    ports:
      - "8084:5000"
    environment:
      - INVENIO_SQLALCHEMY_DATABASE_URI=postgresql+psycopg2://invenio:invenio_secret@postgresql/invenio
      - INVENIO_CACHE_REDIS_URL=redis://redis:6379/0
      - INVENIO_BROKER_URL=amqp://guest:guest@rabbitmq:5672/
      - INVENIO_ELASTICSEARCH_URL=elasticsearch:9200
    depends_on:
      - postgresql
      - redis
      - rabbitmq
      - elasticsearch
    restart: unless-stopped

  postgresql:
    image: postgres:14
    environment:
      - POSTGRES_USER=invenio
      - POSTGRES_PASSWORD=invenio_secret
      - POSTGRES_DB=invenio
    volumes:
      - pg_data:/var/lib/postgresql/data
    restart: unless-stopped

  elasticsearch:
    image: docker.elastic.co/elasticsearch/elasticsearch:8.11.0
    environment:
      - discovery.type=single-node
      - ES_JAVA_OPTS=-Xms512m -Xmx512m
    volumes:
      - es_data:/usr/share/elasticsearch/data
    restart: unless-stopped

  redis:
    image: redis:7
    restart: unless-stopped

  rabbitmq:
    image: rabbitmq:3-management
    restart: unless-stopped

volumes:
  pg_data:
  es_data:

Invenio requires running invenio setup commands after deployment to initialize the database, create indexes, and set up the first administrator user. The full setup process is documented in the InvenioRDM installation guide.

Why Self-Host Your Library Platform?

Self-hosting library management and digital collection software offers significant advantages over commercial SaaS platforms:

Complete data sovereignty: Library catalogs and digital collections represent decades of institutional investment in cataloging, digitization, and curation. Self-hosting ensures this data remains under your control — no vendor can change pricing, discontinue features, or go out of business and take your catalog with them. Your MARC records, digitized items, and metadata exports are always accessible.

Standards compliance on your terms: Commercial platforms often implement library standards (MARC21, Dublin Core, OAI-PMH) incompletely or with proprietary extensions. Open-source platforms let you implement standards exactly as specified, ensuring interoperability with union catalogs, discovery services, and other institutions. You control when to adopt new standards like BIBFRAME or linked data profiles.

Custom public interfaces: Library websites are the primary point of access for patrons. Self-hosted platforms let you customize the OPAC, discovery interface, and exhibit pages to match your institutional brand and accessibility requirements. You can integrate with your library’s existing website, authentication systems, and discovery layers.

Long-term preservation: Digital collections require decades-long preservation commitments. Open-source platforms with active communities (Koha since 1999, Omeka since 2006, Invenio since 2000) provide stability that proprietary platforms cannot guarantee. Your institution controls the upgrade schedule and can maintain compatibility with legacy formats as long as needed.

For institutions managing digital archives and preservation, our digital archive comparison evaluates Archivematica, Omeka S, and DSpace for long-term digital preservation workflows. If you need ebook and audiobook server capabilities alongside your ILS, our ebook server guide covers Audiobookshelf, Kavita, and Calibre-Web for personal and small library media collections.

Choosing the Right Platform

FAQ

What is the difference between an ILS and a digital repository?

An Integrated Library System (ILS) like Koha manages physical and digital library operations — cataloging, circulation, patron management, and acquisitions. A digital repository like Invenio stores, preserves, and provides access to digital objects — research papers, datasets, images, and archival materials. They serve different purposes but can complement each other in a library ecosystem.

Can I use Omeka S as a library catalog?

Omeka S is not designed as a library catalog. It lacks circulation management, MARC cataloging, patron accounts, and barcode integration. It excels at creating curated digital exhibits and collections. For a library that needs both catalog management and digital exhibitions, run Koha (for the catalog) alongside Omeka S (for exhibits).

Do these platforms support multi-language catalogs?

Yes. Koha supports multi-language OPAC interfaces and MARC records in multiple languages. Omeka S has built-in translation support for interface text and can store metadata in any language. Invenio supports multilingual metadata fields and localized discovery interfaces.

How much server space do I need?

For a medium library (50,000 catalog records):

• Koha: 2 CPU cores, 4 GB RAM, 50 GB storage (catalog data only)
• Omeka S: 2 CPU cores, 4 GB RAM, 100+ GB storage (depends on media file sizes)
• Invenio: 4 CPU cores, 8 GB RAM, 200+ GB storage (Elasticsearch + PostgreSQL + file storage)

File storage for digitized collections is the primary driver of disk requirements.

Can these platforms integrate with existing authentication systems?

Yes. Koha supports LDAP and Shibboleth authentication. Omeka S supports SAML via modules. Invenio supports OAuth2, SAML, and LDAP through the Invenio-accounts module. All three can integrate with institutional single sign-on (SSO) systems.

Is there commercial support available?

Yes. Koha has multiple commercial support providers worldwide (ByWater Solutions, Catalyst, etc.). Omeka S is supported by developers at the RRCHNM and a network of consultants. Invenio is backed by CERN and offers commercial support through InvenioRDM partnerships with universities and research organizations.