Firestore is Google Cloud's serverless document database for mobile, web and AI apps, with offline sync, vector search and a MongoDB-compatible edition.
Firestore keeps an application's data as documents grouped in collections and sends each change to the phones and browsers listening for it. Google builds and runs it on Google Cloud as a fully managed service with no servers to provision. It is the database behind the Firebase client SDKs, and the same service also runs in Datastore mode. Documents take whatever fields an app needs, and Firestore indexes single fields automatically, with composite indexes for queries across several fields. Transactions read and write several documents so that all of the changes apply or none do. Reads are strongly consistent by default, and data is split across storage servers without anyone choosing a shard key. Firestore Security Rules decide what each signed-in user may read or write, and IAM roles govern each database. Vector search finds the nearest neighbors of an embedding, TTL policies delete expired documents, and point-in-time recovery and scheduled backups bring back earlier data. The Enterprise edition adds a MongoDB-compatible API, Pipeline queries with grouping and joins, and text and geospatial search. Firestore suits developers building mobile and web apps, especially teams already signing users in with Firebase Authentication. Game studios use it for player profiles, in-game inventories and game state that stay in step across platforms. Teams adding generative AI features can store embeddings in the same documents as the app data they belong to. The Enterprise edition is aimed at teams that want to bring existing MongoDB application code and drivers onto a serverless Google Cloud service. Both editions are set up from the same Google Cloud account and console. The everyday work is storing user profiles, app state, listings and content for sites and apps that change while people use them. The Android, Apple and web SDKs keep a local copy on the device, so an app keeps working offline and syncs its changes when the connection returns. Backend code running on Cloud Run or Cloud Functions writes to the same database the app reads, and BigQuery and Vertex AI connect to it from the wider Google Cloud platform. It fits teams that want app data, device sync and per-user access rules handled by one managed database.
Pricing
View plans-
Free quota
Free
- One free database per project, in either edition
- Daily quotas reset around midnight Pacific time
- Standard edition, 1 GiB stored, 50,000 document reads, 20,000 writes and 20,000 deletes a day
- 10 GiB of outbound data transfer a month
- Enterprise edition, 50,000 read units, 40,000 write units and 10,000 event units a day, plus 50,000 real-time updates
- TTL deletes, PITR data, backups and restores have no free usage and need billing enabled
- New customers get $300 in credits to spend on Firestore
-
Standard edition
Custom pricing
- Pay as you go at list price, Iowa: $0.03 per 100,000 document reads; prices vary by location
- Document writes $0.09 per 100,000; deletes and TTL deletes $0.01 per 100,000
- Stored data $0.000205479 per GiB-hour, shown as $0.15 per GiB-month
- PITR data $0.15 per GiB-month; backup data $0.03 per GiB-month
- Restore and clone operations $0.20 per GiB
- 1-year committed use discount, reads $0.024 and writes $0.072 per 100,000
- 3-year committed use discount, reads $0.018 and writes $0.054 per 100,000
- Outbound transfer after 10 GiB a month $0.12 per GiB to APAC, Europe and the Americas, $0.11 above 1,024 GiB, $0.08 above 10,240 GiB
- Higher outbound rates to Singapore and Tokyo ($0.14), China ($0.23) and Australia ($0.19) for the first band
- Inbound transfer free; transfer between US regions $0.01 per GiB after 10 GiB a month
- Pipeline operations, text search and MongoDB compatibility need the Enterprise edition
-
Enterprise edition
Custom pricing
- Pay as you go at list price: $0.05 per 1,000,000 read units; prices vary by location
- Write units and managed delete units $0.26 per 1,000,000
- Read units are counted in 4 KiB tranches; full-text search and geospatial queries add read units
- Stored data $0.000328767 per GiB-hour
- Backup data $0.000041096 and PITR data $0.000205479 per GiB-hour
- Restore and clone operations $0.20 per GiB
- Event units $0.10 per 1,000,000; real-time updates $0.30 per 1,000,000
- 1-year committed use discount, read units $0.04, write units $0.208, real-time updates $0.24 per 1,000,000
- 3-year committed use discount, read units $0.03, write units $0.156, real-time updates $0.18 per 1,000,000
- Includes MongoDB compatibility, Pipeline operations and text search
Features
- Included: Schemaless document storage
- Included: Secondary indexes on any field
- Included: Multi-document ACID transactions
- Included: Automatic sharding across nodes
- Included: Tunable consistency per operation
- Not included: Multi-region active-active writes
- Included: Server-side aggregation pipelines
- Not included: Change streams for downstream consumers
- Included: Time-to-live expiry on records
- Included: Integrated full-text search on documents
- Included: Vector search on embedded fields
- Not included: Multi-model data access
- Included: Collection-level access roles
- Not included: Client-side field-level encryption
- Included: Continuous backup with point-in-time restore
- Included: Embedded mobile database with sync
- Not included: Open source community edition
- Included: Serverless on-demand capacity pricing
Additional Features
- Firestore Security Rules per collection path
- IAM roles per database
- Real-time listeners
- Offline persistence in the Android, Apple and web SDKs
- Automatic single-field indexes
- Composite indexes
- Atomic transactions across several documents
- Strongly consistent reads by default
Best for
- Mobile and web app teams building on Firebase
- Game studios storing player profiles, inventories and game state
- Offline-first and collaborative apps that sync across devices
- Developers keeping vector embeddings beside application data
- Teams moving MongoDB code, drivers and integrations to a managed Google Cloud service
- Teams already running their backends on Google Cloud
Use cases
- Storing user profiles and app state for mobile apps
- Keeping app data usable on a device that goes offline
- Building collaborative screens with real-time updates
- Tracking game state and marketplace listings
- Serving personalization data to an app
- Running the content store behind a dynamic website
- Storing embeddings for generative AI retrieval
- Moving a MongoDB application to a serverless database
- Deleting expired documents automatically with TTL policies
- Recovering a database to an earlier point in time
- Searching document text and locations (Enterprise edition)
Screenshots & Videos
Explore Firestore in action
User Reviews
Compare with Similar Software
Click to compare side by side with Firestore
