--- title: Amazon DynamoDB adds native vector search in general... description: Amazon DynamoDB now supports native vector search in general availability across all AWS commercial regions and GovCloud. The feature introduces a new vector... canonical: https://daily.dev/posts/amazon-dynamodb-adds-native-vector-search-in-general-availability-6qzzpil4v twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Amazon DynamoDB adds native vector search in general availability | daily.dev og:description: Amazon DynamoDB now supports native vector search in general availability across all AWS commercial regions and GovCloud. The feature introduces a new vector... og:url: https://daily.dev/posts/amazon-dynamodb-adds-native-vector-search-in-general-availability-6qzzpil4v og:image: https://api.daily.dev/og/posts/6Qzzpil4v.png og:image:alt: Amazon DynamoDB adds native vector search in general availability og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Amazon DynamoDB adds native vector search in general availability **[Collections](https://daily.dev/sources/collections)** · 2 min read · 1 upvotes · 0 comments ## Summary Amazon DynamoDB now supports native vector search in general availability across all AWS commercial regions and GovCloud. The feature introduces a new vector index type and a SearchVectors API supporting Euclidean, Cosine, and Dot product distance functions, with embeddings up to 4096 dimensions. It delivers single-digit millisecond latency at 99%+ recall and is designed to scale to trillions of vectors. The serverless, pay-per-request pricing model applies. This allows teams already using DynamoDB to run semantic search, RAG pipelines, recommendation engines, and agentic memory without maintaining a separate vector database. Key constraints include on-demand capacity mode requirement, eventual consistency for search results, immutable index configuration after creation, and a default limit of 5 vector indexes per table. ## Content Amazon DynamoDB now supports native vector search, available across all commercial AWS Regions and GovCloud. You can store vector embeddings directly alongside your operational data and run similarity searches without maintaining a separate vector database or keeping two systems in sync. ## What it does The feature adds a new vector index type to DynamoDB. You create the index on an attribute that stores embeddings, then query it using the new `SearchVectors` API, which returns ranked results with optional inline filtering. Supported distance functions are Euclidean, Cosine, and Dot product. Embeddings can be up to 4096 dimensions. Latency is single-digit milliseconds at 99%+ recall, and AWS says it's designed to scale to trillions of vectors. The serverless model carries over — no infrastructure to provision, no downtime during scaling, and pay-per-request pricing. ## How to use it Here's the basic setup in Python, using research paper abstracts as an example: 1. Create a DynamoDB table in on-demand capacity mode (required — provisioned mode isn't supported) 2. Define a vector index on the attribute that will hold your embeddings 3. Generate embeddings — Amazon Bedrock's Titan Text Embeddings V2 works here 4. Write items with the embedding stored as a DynamoDB attribute 5. Query with `SearchVectors`, optionally passing filter expressions to narrow results A few configuration details worth knowing: - You can set a custom partition key on the vector index to control how vectors are distributed, which affects both cost and performance at scale - Projection settings let you control which attributes come back with search results - Index configuration is immutable after creation, so get it right before you write data - Default limit is 5 vector indexes per table ## Pricing Charges apply per byte for vector writes, searches, and storage, with a 1 KB minimum per operation. It follows the same pay-per-request model as the rest of DynamoDB. ## What this is good for The main appeal is consolidation. If you're already using DynamoDB for operational data and you need semantic search, RAG pipelines, recommendation engines, or agentic memory, you no longer have to run a separate vector store and keep it synchronized. Anomaly detection and product similarity search are other common cases. One thing to keep in mind: vector search results are eventually consistent, not strongly consistent. That's worth factoring in if your use case requires reading your own writes immediately. ## Questions this post answers ### Does DynamoDB now support vector search natively without a separate vector database? Yes, DynamoDB has a native vector index type with a SearchVectors API, available generally across all commercial AWS Regions and GovCloud. It supports Euclidean, Cosine, and Dot product distance functions, embeddings up to 4096 dimensions, single-digit millisecond latency at 99%+ recall, and eliminates the need to sync data to a separate vector store. _Weigh consolidating vector search into dynamodb against a dedicated vector store on daily.dev._ ### What capacity mode does DynamoDB require to use vector search? Vector indexes require the table to be in on-demand capacity mode; provisioned capacity mode is not supported. Index configuration is also immutable once created, and each table defaults to a limit of 5 vector indexes, so these settings need to be decided correctly before writing data. _Developers planning a dynamodb vector index migration can track setup constraints like this on daily.dev._ ### Are DynamoDB vector search results strongly consistent? No, vector search results are eventually consistent rather than strongly consistent. This matters for applications that need to read their own writes immediately after inserting new embeddings, such as certain RAG or agentic memory pipelines. _Anyone designing read-after-write guarantees for rag pipelines can follow consistency nuances like this on daily.dev._ ## Similar posts on daily.dev - [AWS Introduces Native Vector Search for DynamoDB](https://daily.dev/posts/aws-introduces-native-vector-search-for-dynamodb-nn2cx8i4z) · InfoQ · 0 upvotes · 0 comments - [Amazon S3 Vectors now generally available with increased scale and performance](https://daily.dev/posts/amazon-s3-vectors-now-generally-available-with-increased-scale-and-performance-0gpqkus8w) · AWS · 0 upvotes · 0 comments --- Tags: [#aws](https://daily.dev/tags/aws), [#backend](https://daily.dev/tags/backend), [#rag](https://daily.dev/tags/rag), [#vector-search](https://daily.dev/tags/vector-search), [#aws-dynamodb](https://daily.dev/tags/aws-dynamodb) [View this post on daily.dev](https://daily.dev/posts/amazon-dynamodb-adds-native-vector-search-in-general-availability-6qzzpil4v) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Amazon DynamoDB adds native vector search in general availability","url":"https://daily.dev/posts/amazon-dynamodb-adds-native-vector-search-in-general-availability-6qzzpil4v","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/amazon-dynamodb-adds-native-vector-search-in-general-availability-6qzzpil4v"},"datePublished":"2026-08-05T21:42:58.065Z","dateModified":"2026-09-13T19:21:41.335Z","description":"Amazon DynamoDB now supports native vector search in general availability across all AWS commercial regions and GovCloud. The feature introduces a new vector...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/adfd6a103911f529c5a2bbbb2c1bf4e7?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/adfd6a103911f529c5a2bbbb2c1bf4e7?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/amazon-dynamodb-adds-native-vector-search-in-general-availability-6qzzpil4v","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"aws,backend,rag,vector-search,aws-dynamodb","timeRequired":"PT2M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Amazon DynamoDB adds native vector search in general availability"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/amazon-dynamodb-adds-native-vector-search-in-general-availability-6qzzpil4v#faq","mainEntity":[{"@type":"Question","name":"Does DynamoDB now support vector search natively without a separate vector database?","acceptedAnswer":{"@type":"Answer","text":"Yes, DynamoDB has a native vector index type with a SearchVectors API, available generally across all commercial AWS Regions and GovCloud. It supports Euclidean, Cosine, and Dot product distance functions, embeddings up to 4096 dimensions, single-digit millisecond latency at 99%+ recall, and eliminates the need to sync data to a separate vector store. Weigh consolidating vector search into dynamodb against a dedicated vector store on daily.dev."}},{"@type":"Question","name":"What capacity mode does DynamoDB require to use vector search?","acceptedAnswer":{"@type":"Answer","text":"Vector indexes require the table to be in on-demand capacity mode; provisioned capacity mode is not supported. Index configuration is also immutable once created, and each table defaults to a limit of 5 vector indexes, so these settings need to be decided correctly before writing data. Developers planning a dynamodb vector index migration can track setup constraints like this on daily.dev."}},{"@type":"Question","name":"Are DynamoDB vector search results strongly consistent?","acceptedAnswer":{"@type":"Answer","text":"No, vector search results are eventually consistent rather than strongly consistent. This matters for applications that need to read their own writes immediately after inserting new embeddings, such as certain RAG or agentic memory pipelines. Anyone designing read-after-write guarantees for rag pipelines can follow consistency nuances like this on daily.dev."}}]} ```