Vector databases emerged as an essential technology foundation at the beginning of the modern generation AI era.
However, what has changed in the last year is that vectors, the numerical representation of data used by LLM, have increasingly become another data type in all kinds of different databases. Now, Amazon Web Services (AWS) is taking the next leap in ubiquity of vectors with the general availability of Amazon S3 Vector.
Amazon S3 is an AWS cloud object storage service that is widely used by organizations of all sizes to store any and all types of data. Often, S3 is also used as a foundational component for data lake and lakehouse deployments. Amazon S3 Vector now adds native vector storage and similarity search capabilities directly to S3 object storage. Instead of requiring a separate vector database, organizations can store vector embeddings in S3 and query them for semantic search, retrieval-augmented generation (RAG) applications, and AI agent workflows without moving the data to specialized infrastructure.
The service was first previewed in July with an initial capacity of 50 million vectors in a single index. With the GA release, AWS has increased this dramatically to 2 billion vectors in a single index and 20 trillion vectors per S3 storage bucket.
According to AWS, customers created more than 250,000 vector indexes and used more than 40 billion vectors in the four months since the preview launch. The increase in scale with the GA launch now allows organizations to consolidate entire vector datasets into a single index rather than splitting them across the infrastructure. The GA launch also shakes up the enterprise data landscape by providing a new production-ready approach to vectors that could potentially disrupt the market for purpose-built vector databases.
Adding fuel to the competitive fire, AWS Claim S3 Vector Service Can Help Organizations "Reduce the total cost of storing and querying vectors by up to 90% compared to specialized vector database solutions."
AWS S3 positions Vector as a complement, not a competitor, to the Vector database
While Amazon S3 vectors provide a powerful set of vector capabilities, the answer to whether it replaces the need for a dedicated vector database is somewhat nuanced – and it depends on who you ask.
Despite aggressive cost claims and dramatic scale improvements, AWS is positioning S3 Vector as a complementary storage tier rather than a direct replacement for specialized vector databases.
"Customers choose whether they use S3 Vector or Vector Database depending on what the application requires for latency," Mai-Lan Tomsen Bukovec, vice president of technology at AWS, told VentureBeat.
Bukovec said one way to think about it is ‘performance tiering’ based on the organization’s application needs. He said that if the application requires super-fast low-latency response times, a vector database like Amazon OpenSearch is a good choice.
"But for many types of operations, such as building a semantic layer of understanding on your existing data or expanding agent memory with a lot of context, S3 Vector is a great fit."
The question of whether S3 and its low-cost cloud will replace the object storage type is not new to data professionals. Bukovec offered an analogy to how enterprises use data lakes today.
"I expect we will see vector storage evolve similarly to tabular data in data lakes, where customers continue to use transactional databases like Amazon Aurora for certain types of workloads and in parallel use S3 for application storage and analytics, because the performance profile works out and due to data growth they require the S3 traits of durability, scalability, availability, and cost economics."
How customer demand and requirements shaped Amazon S3 vector services
In the first few months of the preview, AWS learned what real enterprise customers really want and need from a vector data store.
"We got very positive feedback from the preview, and customers told us they wanted the capabilities, but at much greater scale and with lower latency, so they could use S3 as the primary vector store for their rapidly growing vector storage," Bukovec said.
In addition to improved scale, query latency for frequently asked queries has dropped to around 100 milliseconds or less, while rare queries complete in less than a second. AWS has increased the maximum search results per query from 30 to 100, and write performance now supports up to 1,000 PUT transactions per second for single-vector updates.
Use cases gaining traction include hybrid search, agent memory extensions, and building semantic layers on existing data.
Bukovec noted that one preview customer, March Networks, uses S3 Vector for large-scale video and photo intelligence.
"The economics of vector storage and latency profile mean that MAR networks can economically store billions of vector embeddings," He said. "Our built-in integration with Amazon Bedrock means it makes it easy to incorporate vector storage into generative AI and video workflows."
Vector database vendors highlight performance gaps
Specialized vector database providers are highlighting significant performance gaps between their offerings and AWS’s storage-centric approach.
Purpose-built vector database providers, including pine nutsOthers, including Viviate, Quadrant, and Chroma, have set up production deployments with advanced indexing algorithms, real-time updates, and purpose-built query optimization for latency-sensitive workloads.
Pinecone does not see Amazon S3 Vector as a competitive challenge to its vector database.
"Before Amazon S3 Vectors first launched, we were actually informed about the project and we did not consider the cost-performance to be directly competitive at scale," Jeff Zhu, vice president of product at Pinecone, told VentureBeat. "This is especially true now with our dedicated read nodes, for example, one of our major e-commerce marketplace customers recently benchmarked a recommended use case with 1.4B vectors and achieved 5.7k QPS at 26ms p50 and 60ms p99."
Analysts divided on vector database future
The launch has revived the debate over whether vector search will remain a standalone product category or become a feature that major cloud platforms commoditize through storage integration.
"It has been clear for some time now that a vector is a feature, not a product," Corey Quinn, chief cloud economist at The Duckbill Group, wrote in a Message on X (formerly Twitter) in response to a question from VentureBeat. "Now everything says this; The rest will follow soon."
Constellation Research analyst Holger Mueller also sees Amazon S3 Vector as a competitive threat to standalone vector database vendors.
"Now it’s back to the vector vendors to make sure they are further and better," Mueller told VentureBeat. "In enterprise software, suites always win."
Muller also highlighted the benefit of AWS’s approach to eliminating data movement. He said that vectors are a means for LLMs to understand enterprise data. The real challenge is how to build the vectors, including how and how often the data is transferred. By adding vector support to S3, where large amounts of enterprise data are already stored, the data movement challenge can be solved.
"CXOs like this approach, because creating vectors requires no data movement," Muller said.
Gartner Distinguished VP Analyst Ed Anderson sees growth for AWS with new services, but he doesn’t expect this to spell the end of Vector Database. He said organizations using S3 for object storage can increase their use of S3 and potentially eliminate the need for dedicated vendor databases. This will increase value for S3 customers while increasing their reliance on S3 storage.
Even with that growth potential for AWS, vector databases are still essential, at least for now.
"Amazon S3 Vector will be valuable to customers, but will not eliminate the need for Vector databases, especially when use cases require low-latency, high-performance data services," Anderson told VentureBeat.
AWS itself appears to be adopting this complementary approach, indicating continued performance improvements.
"We are just getting started on both scale and performance for S3 Vector," Bukovec said. "Just as we’ve improved data read and write performance in S3 for everything from videos to Parquet files, we’ll do the same for vectors."
What does this mean for enterprises
Beyond the debate over whether vector databases survive as standalone products, enterprise architects face immediate decisions about how to deploy vector storage for production AI workloads.
The performance tiering framework provides a clear decision path for enterprise architects evaluating vector storage options.
S3 Vectors works for workloads that tolerate 100ms latency: semantic search on large document collections, agent memory systems, batch analysis on vector embeddings, and background RAG context-retrieval. The economics have become attractive at scale for organizations already invested in AWS infrastructure.
Specialized vector databases remain essential for latency-sensitive use cases: real-time recommendation engines, high-throughput search serving thousands of concurrent queries, interactive applications where users wait synchronously for results, and workloads where performance stability outweighs cost.
For organizations running both workload types, a hybrid approach reflects how enterprises already use data lakes, deploying specialized vector databases for performance-critical queries while using S3 vector for large-scale storage and less time-sensitive operations.
The key question is not whether to replace existing infrastructure, but rather how to provision vector storage at performance levels based on workload requirements.
<a href