AWS announced that Amazon Simple Storage Service (Amazon S3) is now available directly through Vercel Marketplace and v0 by Vercel, enabling web and AI application developers to create storage resources and connect them to their projects from the Vercel dashboard, instead of switching to the AWS Management Console to configure policies and credentials.
The integration covers three types of use: general-purpose S3 buckets, Amazon S3 Tables for analytics, and Amazon S3 Vectors for storing and querying vectors in applications such as semantic search and retrieval-augmented generation (RAG).
Three Storage Paths
General-purpose S3 buckets are intended for storing files, images, videos, documents, and data lake applications. AWS says the service is designed for 99.999999999% data durability, with a default availability of 99.99%.
S3 Tables provides fully managed Apache Iceberg tables, with automatic maintenance and compaction to improve query performance as data grows. They can be queried from engines compatible with the Iceberg REST Catalog interface, including DuckDB.
S3 Vectors provides dedicated vector storage and querying, with costs for uploading, storing, and querying them reduced by up to 90%, according to AWS. However, the service does not create the embeddings itself; developers must generate them using the model of their choice.
What Changes Practically for Developers?
From the Vercel dashboard, a developer selects Storage and then Create, followed by AWS, the S3 bucket type, and the region, with provisioning completed within seconds. The bucket can be connected to a project, and environment variables can be pulled through the Vercel CLI.
The integration relies on the awsCredentialsProvider provider in @vercel/functions/oidc to exchange a short-lived OIDC token with AWS STS and obtain temporary, resource-constrained credentials. This means static AWS keys are not stored inside the project or platform.
Integration with v0 and Other AWS Services
v0 can interpret natural-language descriptions such as creating a photo gallery or a support bot that answers questions about PDF files, then select the appropriate S3 resource type, configure the connection, and generate runnable code. S3 can also be integrated with AWS services available in Vercel, such as Aurora, DynamoDB, and OpenSearch Serverless, within the same account and authentication model.
The platform allows users to create a new AWS account that supports S3 features for building and testing, or connect an existing account while retaining its current billing. However, when an existing account is connected, the integration works only with new buckets created through Vercel.
Availability and Cost
Vercel does not charge fees for the integration itself, while resources are billed according to standard Amazon S3 pricing. New AWS accounts created through Vercel receive credits worth $100, valid for six months. Amazon S3, including S3 Tables and S3 Vectors, is available through v0 and Vercel Marketplace in all commercial AWS regions supported by Vercel.
Editorial reading: The core value of the announcement is not the launch of a new storage type, but the transfer of part of backend infrastructure setup to the Vercel workflow while keeping resources inside the user's AWS account. This reduces the initial friction for web and AI teams, but it does not eliminate responsibility for choosing regions, managing costs, and relying on AWS services. S3 Vectors also requires developers to provide the embedding-generation process themselves.