RDS Supabase is a managed Supabase service provided by RDS for PostgreSQL. You can use this service to quickly build backend services on top of RDS for PostgreSQL, ideal for use cases such as minimum viable product (MVP) validation, AI Application Development, and SaaS platforms.
Features
What is Supabase?
Supabase is an open-source Backend-as-a-Service (BaaS) platform designed to provide developers with features similar to Firebase. Supabase is fully open-source and supports self-hosting. Built on PostgreSQL, the platform provides core services such as a database, authentication, file storage, and Realtime capabilities, making it ideal for developers who need to rapidly build an application backend.
Supabase provides a complete backend toolkit for building applications, allowing you to use the entire suite or select only the components you need.
The core capabilities of Supabase include a database, authentication and authorization, object storage, edge functions, Realtime, vectors, and artificial intelligence (AI).
Core capabilities of RDS Supabase
RDS Supabase provides the core features of open-source Supabase, optimized to run on the cloud-native infrastructure of RDS for PostgreSQL.
Database
Provides standard PostgreSQL capabilities and additional extensions. It also supports proprietary RDS for PostgreSQL extensions, such as using Supabase with the rds_ai extension to build AI applications.
Supports auto-generated Data APIs, including CRUD, RESTful, and GraphQL APIs.
Authentication and authorization
Provides a complete user management system that supports multiple authentication methods and can be integrated separately.
Object storage
Provides direct integration with the RDS Cold Storage service, which is built on Alibaba Cloud OSS.
Fully compatible with the Alibaba Cloud OSS protocol, allowing you to seamlessly integrate your application with OSS.
Supports advanced features such as resumable uploads and storage compression.
Realtime
When data changes, updates are synchronized to all connected clients in real time.
Stores and synchronizes the online status of all users across all clients.
Supports channel subscriptions and broadcasting.
Vectors and AI
Vector service
Uses the built-in rds_pgvector extension to store, index, and access embeddings.
Storing vectors with transactional data in the same database simplifies your application and improves performance.
API integration
Easily connects to any large language model (LLM) or embeddings API, including Hugging Face and SageMaker.
AI assistant
Includes a built-in Model Studio (Bailian) token to provide LLM services directly through Qwen.
Use cases
Minimum viable product (MVP) validation
Helps startups and independent developers release a viable application sooner for a faster time-to-market.
After designing your prototype, you can quickly generate backend services with simple table configurations.
Quickly integrate features like Supabase authentication and message subscriptions to significantly improve development efficiency.
AI Application Development
By combining RDS Supabase with PostgreSQL, you get an all-in-one solution for developing AI applications.
RDS for PostgreSQL provides multi-modal data storage and retrieval capabilities based on the rds_pgvector extension.
RDS Supabase provides API endpoints for your application.
You can use the rds_ai capability to quickly connect with model services from Model Studio (Bailian) and PAI.
SaaS platforms
With RDS Supabase, business teams can efficiently develop internal platforms throughout the full lifecycle. This helps enterprises monitor project data in real time, manage content assets collaboratively, and improve cross-departmental efficiency.
Prerequisites
The required service-linked roles (SLRs) are authorized
Authorization is required only on the first use and incurs no fees. The following three service-linked roles must be authorized:
AliyunServiceRoleForRds
AliyunServiceRoleForRdsPgsqlOnEcs
AliyunServiceRoleForRDSAISupabase
You can click Go to Authorization on the page for creating an RDS Supabase project to complete the authorization.
The required permissions are granted to the RAM user
If you use a Resource Access Management (RAM) user to access this feature, the RAM user must have the following permissions. For instructions on granting permissions, see Grant permissions to a RAM user.
AliyunRDSAIReadOnlyAccess or AliyunRDSAIFullAccess: Read-only or full management permissions for AI Application Development in RDS.
AliyunVPCFullAccess: Permissions to manage VPCs.
AliyunRDSFullAccess: Permissions to manage RDS.
AliyunRAMReadOnlyAccess: Read-only permissions for Resource Access Management (RAM), which allows viewing users, groups, and authorization information.
Billing
RDS for PostgreSQL instance fees
When you create an RDS Supabase project, you must create an RDS for PostgreSQL instance within it. You will be charged for the RDS instance type and storage. The instance uses the same billing method as the RDS Supabase project.
The Supabase module is free. You are only charged the RDS for PostgreSQL instance fees on a pay-as-you-go basis.
Usage notes
Currently, only PostgreSQL major version 17 and later are supported.
Do not delete or modify the databases and associated accounts automatically created by the RDS Supabase project. Doing so renders the project unusable.
Create an RDS Supabase project
Go to the RDS console. In the left-side navigation pane, choose AI Capability Center > Agent Development.
Select a region at the top of the page. On the RDS AI Application Platform (formerly RDS Supabase) tab, click Create Project.
On the page that appears, configure the RDS Supabase parameters.
Parameter
Description
RDS AI Application Project Name
Enter a name for the RDS AI Application project.
SLR Authorization
Authorization is required only on the first use and does not incur any fees. Click Go to Authorization to authorize the following service-linked roles: AliyunServiceRoleForRds, AliyunServiceRoleForRdsPgsqlOnEcs, and AliyunServiceRoleForRDSAISupabase.
Region
The region where the RDS AI Application project is located.
VPC
Select the virtual private cloud (VPC) for the RDS AI Application project.
If you have an existing VPC that meets your network planning needs, you can select it directly.
If you do not have a suitable VPC, you can use the default VPC and vSwitch. For more information, see VPC and vSwitch.
If the default VPC and vSwitch do not meet your requirements, create a VPC and vSwitch first. For instructions, see VPC and vSwitch.
Primary zone and network
Select the primary zone and vSwitch.
RDS AI Application Module
Select the RDS AI Application modules to enable. The following modules are supported: supabase, Knowledge Graph, Sandbox and Edge Functions, and mem0.
To enable a Supabase instance, select supabase.
Dashboard Username
Dashboard Password
Set the password for the default user supabase.
The password must contain at least three of the following character types: uppercase letters, lowercase letters, digits, and underscores (_). The length must be 8 to 32 characters.
Billing Method
Subscription and pay-as-you-go are supported.
Engine
The engine version of the RDS for PostgreSQL instance in the RDS AI Application project. PostgreSQL 17 and PostgreSQL 18 are supported.
Edition
Select the edition of the RDS for PostgreSQL instance in the RDS AI Application project.
Basic Edition: A single-node, read-only instance that is cost-effective and suitable for learning or testing. Fault recovery and restarts take a long time.
High-availability Edition: A primary node and a secondary node provide high availability. This edition is suitable for production environments and fits more than 80% of use cases.
Instance Family
Select the instance family of the RDS for PostgreSQL instance in the RDS AI Application project.
General-purpose: Dedicated memory and I/O with shared CPU and storage. Offers a lower price and higher cost-effectiveness.
Dedicated: Dedicated memory, I/O, CPU, and storage. Provides superior and stable performance.
NoteThe Basic Edition does not support dedicated instance families.
Instance Type
Select the instance type for the RDS for PostgreSQL instance in the RDS AI Application project.
NoteFor supported instance types, refer to the information in the console. For more details, see RDS for PostgreSQL instance types.
Storage
Set the storage capacity for the RDS for PostgreSQL instance in the RDS AI Application project.
Database Password
Set the database password for the RDS for PostgreSQL instance in the RDS AI Application project.
The password must contain at least three of the following character types: uppercase letters, lowercase letters, digits, and underscores (_). The length must be 8 to 32 characters.
After confirming the configuration, click Buy Now.
After the project is created, you can view it on the RDS AI Application Platform (formerly RDS Supabase) tab.
Log on to the Supabase console (Supabase Dashboard)
Go to the RDS console. In the left-side navigation pane, choose AI Capability Center > Agent Development.
Select a region at the top of the page. On the RDS AI Application Platform (formerly RDS Supabase) tab, click the ProjectID of the target project to go to the project details page.
On the Basic Information page, in the Allowlist Information section, click Add Allowlist Group to add the IP address of your client to the allowlist.
NoteIf you have already configured an allowlist, you can directly click the Public Endpoint to go to the Supabase logon page.
Click the Public Endpoint in the Network Information section to go to the Supabase logon page.
For more information about using Supabase, see Getting Started | Supabase Docs.
NoteIf you log on from an ECS instance in the same VPC as the RDS Supabase project, we recommend using the Internal Endpoint.
In the Actions column, click View Details. In the Basic Information dialog box that appears, you can view detailed information about the instance.
Delete an RDS Supabase project
Go to the RDS console. In the left-side navigation pane, choose AI Capability Center > Agent Development.
Select a region at the top of the page. On the RDS AI Application Platform (formerly RDS Supabase) tab, click Delete next to the target project.
In the dialog box that appears, click OK.
ImportantDeleting an RDS Supabase project does not automatically delete the RDS for PostgreSQL instance created with the project. You must manually release the instance.
FAQ
References and APIs
RDS Supabase SDK User Guide: Describes how to obtain the Supabase URL and Supabase Key, and how to use the RDS Supabase SDK in various programming languages.
Related APIs For APIs related to RDS Supabase, see Supabase API Reference.
Appendix: Obtain a service API key
Method 1: Use an API
Call DescribeInstanceAuthInfo to obtain the service API key.
Method 2: Use the RDS console
Go to the RDS console. In the left-side navigation pane, choose AI Capability Center > Agent Development.
Select a region at the top of the page. On the RDS AI Application Platform (formerly RDS Supabase) tab, click the ProjectID of the target project to go to the project details page.
In the upper-right corner of the Basic Information page, click Get API Key.
In the ApiKeys dialog box that appears, copy the string next to ServiceKey to obtain the service API key.