FerretDB
At #13 of 22 in our database management systems ranking, FerretDB is summed up in the catalogue as: Open-source proxy that serves MongoDB drivers from PostgreSQL.
Sebastiaan SmitsFounderPaid placement exists here and is disclosed. Whatever the arrangement, the FerretDB link carries the same marker as all others (details).
Visit FerretDB See it in the database management systems ranking
At a glance
| Headquarters | Not stated by the vendor |
|---|---|
| Free plan | Yes, a free tier or plan |
| Pricing model | Free under Apache-2.0; you host PostgreSQL with the DocumentDB extension |
| Best for | Teams keeping MongoDB drivers who want an Apache-licensed server |
| API | Not covered by the sources we used |
| Deployment | Left open by the sources we used |
What is FerretDB?
FerretDB is an open-source proxy from FerretDB Inc. that lets applications written for MongoDB talk to PostgreSQL. It accepts the MongoDB 5.0+ wire protocol, translates queries to SQL, and stores documents in PostgreSQL with the DocumentDB extension. Applications keep their existing MongoDB drivers and connection strings; only the server behind them changes. The project's stated reason for existing is licensing: MongoDB moved to the SSPL, which the FerretDB docs say made it unsuitable for many open-source and early-stage commercial projects, while FerretDB ships under Apache-2.0.
The vendor advises checking suitability before production and keeps a public list of known differences and supported commands. You can run it yourself from Docker images, DEB packages or Kubernetes, or use FerretDB Cloud, a managed service. It suits teams that like the MongoDB developer experience but want an open-source licence and PostgreSQL underneath for storage, backups and operations.
FerretDB key features
- MongoDB wire compatibility. Drivers and applications built for MongoDB 5.0+ are meant to work unchanged. The docs list compatible applications tested by the team, and a compatibility page records every command as supported or not yet implemented, so gaps can be checked before a move.
- PostgreSQL with DocumentDB storage. Data lives in PostgreSQL with the DocumentDB extension, so the usual PostgreSQL tools apply to the same instance. The evaluation image bundles both pieces plus the MongoDB Shell, and you can connect with psql as well as mongosh.
- Authentication through PostgreSQL. FerretDB stores no credentials itself. It forwards usernames and passwords to PostgreSQL, which validates them, and only SCRAM-SHA-256 is supported on the client side. Users can be created in PostgreSQL directly or through the createUser command.
- TLS connections. Start the server with listen-tls, certificate, key and CA file flags or their environment variables, then add tls parameters to the MongoDB URI. The CA file option also supports verification of client certificates.
- Search and indexing guides. The guides section covers replication, vector search, full-text search and time-to-live indexes, each explaining the parameters FerretDB supports. These help teams check whether features such as embeddings or expiring documents are available before they migrate.
- Migration tooling. The migration section has pre-migration testing and a MongoDB guide that uses native tools: mongodump and mongorestore, or mongoexport and mongoimport. A separate page covers moving from FerretDB v1.x to v2.
- FerretDB Cloud. A managed service where you choose cloud provider, region, network type and compute size when creating a deployment. The docs list four plans: Free, Pro, Enterprise, and Bring Your Own Account Enterprise. Subscription requests are approved by email.
Who FerretDB suits
- A start-up whose CMS expects MongoDB. Payload CMS, NodeBB and Form.io appear on the compatible applications list. The team can point those applications at FerretDB and keep one PostgreSQL service for storage, rather than adding a separate MongoDB cluster.
- A platform team that standardises on PostgreSQL. CloudNativePG is listed among deployment platforms, so Kubernetes operators already running PostgreSQL can host the DocumentDB-enabled database. Backups, monitoring and access control then follow existing PostgreSQL practice.
- A developer evaluating a MongoDB replacement. A single docker run command starts the evaluation image with FerretDB, PostgreSQL and mongosh. The developer runs real queries, then checks the known-differences list and the pre-migration testing page before touching production data.
Getting started with FerretDB
For a quick trial the docs give one Docker command that starts the evaluation image, ghcr.io/ferretdb/ferretdb-eval:2, with FerretDB, PostgreSQL with DocumentDB and the MongoDB Shell; you set a username and password and connect on port 27017. For real deployments there are separate installation guides for FerretDB (Docker and packages) and for DocumentDB (Docker, Kubernetes, DEB).
To move data, run the pre-migration test first, then export with mongodump or mongoexport and restore into FerretDB. The learning curve is mostly PostgreSQL operations, since FerretDB depends on the extension. FerretDB Cloud has an account sign-up with email verification.
What we make of FerretDB
FerretDB translates MongoDB 5.0+ wire protocol into SQL and keeps documents in PostgreSQL with the DocumentDB extension. The licence is Apache-2.0, so the SSPL question disappears. The docs say to verify suitability before production and point to a public list of known differences.
FerretDB is entry #13 of 22 in the database management systems list: it follows Amazon DynamoDB and comes before RavenDB. Challengers hold places 1 to 3 where a category has them; the number here is a place in the order, not a mark out of ten.
The vendor states no country of establishment, so FerretDB is left out of the regional count for this list.
Where FerretDB wins and where it does not
- FerretDB translates MongoDB 5.0+ wire protocol into SQL and keeps documents in PostgreSQL with the DocumentDB extension.
- Strongest for: Teams keeping MongoDB drivers who want an Apache-licensed server.
- MongoDB wire protocol is among the traits we single out for it.
- Catalogue tag: Runs on PostgreSQL.
- Marked in our records as apache-2.0.
- Compatibility is partial and documented as known differences, so queries need testing.
- Not for: Teams that rely on Atlas Search or Vector Search features.
- The docs list no managed service, so you operate PostgreSQL yourself.
FerretDB pros and cons in practice
- Apache-2.0 licensing removes the SSPL question, and the code is public, so you can audit how queries translate to SQL.
- Existing MongoDB drivers, GUI clients and many applications connect without code changes, which lowers the cost of a trial.
- Using PostgreSQL underneath means one storage engine, with familiar backup and monitoring tools.
- Compatibility is documented command by command, with a public roadmap, so gaps are visible before migration.
- It is compatible with MongoDB 5.0+ in many cases, not all; the docs tell you to verify suitability and list known differences.
- Error messages differ from MongoDB's, and collection names must be valid UTF-8, differences the project says it will not fix.
- Authentication supports only SCRAM-SHA-256 on the client side and relies entirely on PostgreSQL users.
- Cloud plan prices are not shown in the documentation, so budgeting for the managed route needs a vendor conversation.
Paying for FerretDB
Pricing as recorded: Free under Apache-2.0; you host PostgreSQL with the DocumentDB extension.
The software is free: the repository carries the Apache-2.0 licence, and you pay only for the PostgreSQL hosting you run. FerretDB Cloud is the paid route. The docs name four plans, Free Tier for developers, students and small projects, Pro Tier for teams wanting predictable pricing, Enterprise Tier with dedicated support, and Bring Your Own Account Enterprise Tier. No amounts, billing periods or usage limits are printed in the documentation we fetched, and the vendor's pricing address returned not found, so prices are not stated here. Support beyond community channels (Slack, GitHub Discussions and Issues) is tied to the Cloud and Enterprise plans.
Read from docs.ferretdb.io when Best Database Management Systems in 2026 was last reviewed (7 November 2026).
Living with FerretDB: setup and exit
The docs describe a proxy that accepts MongoDB 5.0 and later protocol traffic and runs on PostgreSQL with the DocumentDB extension.
Source is Apache-2.0 on GitHub.
Copyright is held by FerretDB Inc.; no country is stated in the docs.
The project keeps a roadmap, a list of supported commands and a list of known differences, which is the checklist to run before moving.
FerretDB: data and security
When self-hosted, data sits in your own PostgreSQL, so region, backups and encryption at rest follow your setup. In FerretDB Cloud you pick the cloud provider and region at deployment. The docs cover authentication through PostgreSQL and TLS with optional client certificates.
The telemetry page says the product reports anonymous usage data (instance UUID, versions, command names, not argument values or personal data) and writes the same data to a telemetry.json file you can read. Certifications are not stated on the pages we fetched.
Fact sheet: FerretDB
| Product | FerretDB |
|---|---|
| Vendor website | docs.ferretdb.io |
| Catalogue tags | Apache-2.0, MongoDB wire protocol, Runs on PostgreSQL |
| Head-to-head pages | Not paired so far |
| Rank there | #13 of 22 |
| Gaps in our records | headcount, funding and customer count |
| Primary guide | Best Database Management Systems in 2026 |
| Price visibility | A price or free offer is shown |
| Alternatives page | No dedicated page |
| Headquarters | Not stated by the vendor |
What FerretDB connects to
Integrations the vendor names (12):
- MongoDB Compass
- Studio 3T
- DBeaver
- NoSQLBooster
- CloudNativePG
- Grafana
- OpenTelemetry
- Payload CMS
- NodeBB
- LibreChat
- GrowthBook
- HashiCorp Vault
Also in our records:
- The docs describe a proxy that accepts MongoDB 5.0 and later protocol traffic and runs on PostgreSQL with the DocumentDB extension.
What to look at instead of FerretDB
These are the nearest neighbours of FerretDB in the same list, picked on shared traits and price model.
- Couchbase (#8 of 22): Not for: buyers who want fully published self-managed prices up front. JSON document database with SQL++ queries, sold as Capella or as Server. (United States)
- RavenDB (#14 of 22): Not for: teams that need MongoDB wire compatibility. ACID document database with self-built indexes, sold on-prem and as RavenDB Cloud. (Israel)
- ScyllaDB (#15 of 22): Wide-column database compatible with Cassandra and DynamoDB APIs. Best for: Teams with very high write throughput and key-based access.
- Azure Cosmos DB (#18 of 22): Not for: teams that need the whole MongoDB feature set. Microsoft's distributed database with NoSQL, MongoDB, Cassandra and PostgreSQL APIs. (United States)
- DuckDB (#20 of 22): Not for: teams needing a managed, multi-user server out of the box. Embedded analytical SQL database that runs inside your application process.
Where else FerretDB is ranked
| Guide | Position |
|---|---|
| Best Database Management Systems in 2026 | #13 of 22 |
FerretDB FAQ
9 answersWhich country is FerretDB from?
The vendor pages we reviewed state no country of establishment for FerretDB, so we do not give one.
How much does FerretDB cost?
Our catalogue records FerretDB’s pricing as: Free under Apache-2.0; you host PostgreSQL with the DocumentDB extension. On free access, it has a free tier or free plan.
Which tools compete with FerretDB?
The nearest neighbours in our ranking are Couchbase, RavenDB, ScyllaDB, Azure Cosmos DB and DuckDB. All of them are reviewed in the same ranking.
Is FerretDB one of the top database management systems?
FerretDB is #13 of 22 in our database management systems ranking.
Who is FerretDB best for?
Teams keeping MongoDB drivers who want an Apache-licensed server. It does not fit as well for teams that rely on Atlas Search or Vector Search features.
Is FerretDB a replacement for MongoDB?
The docs call it an open-source alternative that works as a direct replacement for MongoDB 5.0+ in many cases.
It is not identical: a list of known differences and supported commands is public, and the vendor asks you to verify suitability for your application before production use.
What database does FerretDB store data in?
PostgreSQL with the DocumentDB extension. FerretDB converts MongoDB wire protocol queries to SQL and relies on the extension as its engine. The evaluation Docker image packages both, and DocumentDB can also be installed through Docker, Kubernetes or a DEB package.
How do I migrate from MongoDB?
Run the pre-migration testing steps, then export data with mongodump or mongoexport using your existing connection URI, and load it into FerretDB with mongorestore or mongoimport. The docs note mongoexport handles one collection at a time, while mongodump can take the whole instance.
Does FerretDB collect telemetry?
Yes, by default it reports anonymous data such as a random instance UUID, versions, installation type and command names.
The docs say argument values, field names and personal data are never collected, raw data is not sold or shared, and the same content is saved locally in telemetry.json.
FerretDB: sources
Read from the vendor’s own pages on 5 November 2026.
- docs.ferretdb.io: what FerretDB is, SSPL rationale, scope
- docs.ferretdb.io/installation/evaluation: evaluation Docker image and connection
- docs.ferretdb.io/installation/ferretdb-cloud: FerretDB Cloud sign-up and four plans
- docs.ferretdb.io/security/authentication: authentication via PostgreSQL, SCRAM-SHA-256
- docs.ferretdb.io/security/tls-connections: TLS flags
- docs.ferretdb.io/compatible-applications: tested applications and tools
- docs.ferretdb.io/migration/migrating-from-mongodb: migration with native tools
- docs.ferretdb.io/migration/compatibility: known differences
- docs.ferretdb.io/telemetry: telemetry content
- docs.ferretdb.io/guides: replication, vector search, full-text search, TTL indexes
- raw.githubusercontent.com/FerretDB/FerretDB/main/LICENSE: Apache-2.0 licence
How we rank
The same four questions are asked of every database management systems product: time to value, real cost, the way out and the intended buyer. The result for FerretDB: #13, out of 22 ranked. Method and commercial relationships are in our editorial process.