# VectorAI DB

## Docs

- [AI recipe recommendation agent](https://docs.vectoraidb.actian.com/academy/articles/AI-recipe-recommendation-agent.md): Build an AI-powered recipe recommendation agent using Actian VectorAI DB that matches user cravings to recipes through semantic search, filters by dietary restrictions and available ingredients, and learns preferences over time.
- [Multivector document intelligence with visual RAG](https://docs.vectoraidb.actian.com/academy/articles/Multivector-Document-Intelligence-with-Visual-RAG.md): Build a multimodal document intelligence system that embeds PDF pages as images with CLIP, retrieves them via Actian VectorAI DB, and generates answers using GPT-4o vision.
- [Next-gen product discovery with multimodal AI](https://docs.vectoraidb.actian.com/academy/articles/Next-Gen-Product-Discovery-with-Multimodal-AI.md): Build a multimodal hybrid search system combining CLIP dense embeddings and BM25 sparse scoring for semantic and keyword product retrieval using Actian VectorAI DB.
- [AI legal contract intelligence agent](https://docs.vectoraidb.actian.com/academy/articles/building-a-scalable-agent-memory-with-Actian-vector-AI-database.md): Build an AI-powered legal contract analysis system using Actian VectorAI DB with cross-collection lookup, payload-sorted retrieval, connection pooling, and quantization-aware search.
- [Overview](https://docs.vectoraidb.actian.com/academy/articles/index.md): Browse and choose from deep-dive articles on building AI agents and intelligent applications with Actian VectorAI DB.
- [AI supply chain inventory risk intelligence agent](https://docs.vectoraidb.actian.com/academy/articles/supply-chain-inventory-management-agent.md): Build an AI-powered supply chain risk intelligence workflow using Actian VectorAI DB, semantic retrieval, payload filters, and lightweight reasoning.
- [Overview](https://docs.vectoraidb.actian.com/academy/index.md): Learn Actian VectorAI DB through hands-on tutorials, deep-dive articles, and ready-to-run examples.
- [Adaptive RAG systems](https://docs.vectoraidb.actian.com/academy/tutorials/adaptive-rag.md): Learn how to build a Retrieval-Augmented Generation system that adapts its retrieval strategy at runtime based on query type, confidence signals, and user feedback—using Actian VectorAI DB's multistage prefetch, fusion, score thresholds, payload-driven routing, and feedback loops.
- [Build your first application](https://docs.vectoraidb.actian.com/academy/tutorials/first-application.md): Step-by-step guide to building a semantic search app with Actian VectorAI DB: install, connect, create collections, embed, store, search, filter, update, and delete.
- [Overview](https://docs.vectoraidb.actian.com/academy/tutorials/index.md): Hands-on tutorials to build vector search applications with Actian VectorAI DB.
- [Use open-source embedding models](https://docs.vectoraidb.actian.com/academy/tutorials/leverage-open-source-embedding-models.md): Learn how to choose, configure, and integrate open-source embedding models with Actian VectorAI DB—covering model selection, dimensionality trade-offs, distance metrics, batch ingestion, quantization for large models, named vectors for multimodel search, and re-embedding workflows.
- [Building a multimodel system](https://docs.vectoraidb.actian.com/academy/tutorials/multimodel-system.md): Work with multiple embedding models in one system
- [Predicate filters](https://docs.vectoraidb.actian.com/academy/tutorials/predicate-filters.md): Learn how to combine vector similarity search with structured payload filters using Actian VectorAI DB's type-safe Filter DSL.
- [Reranking search results](https://docs.vectoraidb.actian.com/academy/tutorials/re-ranking.md): Learn how to improve search relevance in Actian VectorAI DB by reranking initial retrieval results using multistage prefetch pipelines, cross-encoder scoring, payload-based boosting, fusion reranking, and score thresholds.
- [Optimizing retrieval quality](https://docs.vectoraidb.actian.com/academy/tutorials/retrieval-quality.md): Learn how to measure and improve similarity search accuracy in Actian VectorAI DB by tuning HNSW parameters, choosing the right distance metric, configuring quantization, using multistage prefetch, adjusting score thresholds, and leveraging payload indexes.
- [Similarity search fundamentals](https://docs.vectoraidb.actian.com/academy/tutorials/similarity-search.md): Learn the core vector similarity search workflow with Actian VectorAI DB: embedding, storing, searching, scoring, tuning, batching, and paginating results.
- [Create access token](https://docs.vectoraidb.actian.com/api-reference/access-tokens/create-access-token.md): Creates a new access token with the specified name, description, expiration, and permissions.
- [Delete access token](https://docs.vectoraidb.actian.com/api-reference/access-tokens/delete-access-token.md): Deletes an access token by its ID. The token is immediately invalidated and can no longer be used for authentication.
- [List access tokens](https://docs.vectoraidb.actian.com/api-reference/access-tokens/list-access-tokens.md): Returns a list of all access tokens. The response does not include the raw token values.
- [Rotate access token](https://docs.vectoraidb.actian.com/api-reference/access-tokens/rotate-access-token.md): Generates a new raw token for an existing access token and invalidates the previous raw token immediately. The token `id`, `name`, `description`, `permission`, `will_expire`, and `expired_at` values are preserved.
- [Admin login](https://docs.vectoraidb.actian.com/api-reference/admin-user/admin-login.md): Authenticates the admin user and returns a JWT token. The username is always fixed to `admin`. No authorization header is required for this endpoint.
- [Check admin exists](https://docs.vectoraidb.actian.com/api-reference/admin-user/check-admin-exists.md): Returns whether the admin user has been created. No authorization is required for this endpoint.
- [Create admin user](https://docs.vectoraidb.actian.com/api-reference/admin-user/create-admin-user.md): Creates the built-in admin user with the given password. The username is always fixed to `admin`.
- [Reset admin password](https://docs.vectoraidb.actian.com/api-reference/admin-user/reset-admin-password.md): Resets the built-in admin user's password. The username is always fixed to `admin`.
- [Set auth enabled](https://docs.vectoraidb.actian.com/api-reference/admin-user/set-auth-enabled.md): Enables or disables authentication enforcement for the server. When auth is disabled, endpoints that normally require access tokens or JWT authorization can be called without credentials.
- [Error codes](https://docs.vectoraidb.actian.com/api-reference/error-codes.md): HTTP status codes and error responses returned by the VectorAI DB REST API.
- [gRPC endpoints](https://docs.vectoraidb.actian.com/api-reference/grpc/index.md): gRPC endpoints for VectorAI DB.
- [REST API overview](https://docs.vectoraidb.actian.com/api-reference/rest.md): Complete REST API documentation for VectorAI DB operations.
- [Check if collection exists](https://docs.vectoraidb.actian.com/api-reference/rest/collections/collections/check-if-collection-exists.md): Check whether a collection with the given name exists in the database.
- [Create collection](https://docs.vectoraidb.actian.com/api-reference/rest/collections/collections/create-collection.md): Create a new collection with a specified vector configuration. You must provide the vector dimension and distance metric. Supported distance metrics are `Cosine`, `Euclid`, `Dot`, and `Manhattan`.
- [Delete collection](https://docs.vectoraidb.actian.com/api-reference/rest/collections/collections/delete-collection.md): Delete a collection and all its data permanently. This operation is irreversible. All vectors, payloads, and indexes will be permanently removed.
- [Get collection info](https://docs.vectoraidb.actian.com/api-reference/rest/collections/collections/get-collection-info.md): Get detailed information about a specific collection, including vector configuration, index type and parameters, point count, and collection status.
- [List all collections](https://docs.vectoraidb.actian.com/api-reference/rest/collections/collections/list-all-collections.md): Get a list of all existing collection names in the database. Use `GET /collections/{collection_name}` to get detailed information about a specific collection.
- [Update collection](https://docs.vectoraidb.actian.com/api-reference/rest/collections/collections/update-collection.md): Update parameters of an existing collection, such as optimizer configuration and HNSW search settings. Vector dimension and distance metric cannot be changed after creation.
- [Filter examples](https://docs.vectoraidb.actian.com/api-reference/rest/filters/filters/filter-examples.md): Filters narrow the results of search, scroll, count, and other point operations by applying conditions to payload fields. You pass a filter object in the `filter` parameter of these operations.
- [Clear payload](https://docs.vectoraidb.actian.com/api-reference/rest/points/points/clear-payload.md): Remove all payload fields from the specified points, leaving them with an empty payload. Only explicitly listed point IDs are supported; filter-based targeting is not available for this operation. Points that already have an empty payload are unaffected.
- [Delete payload keys](https://docs.vectoraidb.actian.com/api-reference/rest/points/points/delete-payload-keys.md): Remove specific payload fields from the specified points. Only explicitly listed point IDs are supported; filter-based targeting is not available for this operation. If a key does not exist on a given point, it is silently ignored.
- [Delete points](https://docs.vectoraidb.actian.com/api-reference/rest/points/points/delete-points.md): Delete points from a collection by specifying a list of point IDs.
- [Get points by IDs](https://docs.vectoraidb.actian.com/api-reference/rest/points/points/get-points-by-ids.md): Retrieve multiple points by their IDs. Returns the matching points with their vectors and payloads based on the request parameters.
- [Get single point](https://docs.vectoraidb.actian.com/api-reference/rest/points/points/get-single-point.md): Retrieve full information for a single point by its ID, including its vector and payload.
- [Overwrite payload](https://docs.vectoraidb.actian.com/api-reference/rest/points/points/overwrite-payload.md): Replace the entire payload on the specified points. All existing payload fields are removed and replaced with the provided object. Use the set payload endpoint to merge instead.
- [Set payload](https://docs.vectoraidb.actian.com/api-reference/rest/points/points/set-payload.md): Merge payload fields onto existing points identified by ID. Existing payload fields that are not specified in the request are preserved.
- [Update vectors](https://docs.vectoraidb.actian.com/api-reference/rest/points/points/update-vectors.md): Update vector data for existing points without modifying their payloads. This is useful when re-embedding content or correcting vector values.
- [Upsert points](https://docs.vectoraidb.actian.com/api-reference/rest/points/points/upsert-points.md): Insert or update points in a collection. If a point with the given ID already exists, it is overwritten. All points in the request are inserted or updated atomically.
- [Count points](https://docs.vectoraidb.actian.com/api-reference/rest/search/count/count-points.md): Count the number of points in a collection, optionally filtered by payload conditions. Useful for checking collection size, validating data loads, or counting points matching specific criteria.
- [Scroll points](https://docs.vectoraidb.actian.com/api-reference/rest/search/scroll/scroll-points.md): Paginate through all points in a collection. The response includes a `next_page_offset` cursor that you pass as `offset` in subsequent requests to retrieve the next page. Useful for exporting data, iterating large datasets, or processing points in batches.
- [Batch search](https://docs.vectoraidb.actian.com/api-reference/rest/search/search/batch-search.md): Execute multiple search queries in a single request. This is more efficient than sending individual search requests when you have several queries to run.
- [Search vectors](https://docs.vectoraidb.actian.com/api-reference/rest/search/search/search-vectors.md): Find the most similar vectors in a collection using approximate nearest neighbor search. Supports score thresholds, payload filtering, and tunable HNSW parameters.
- [Collections overview](https://docs.vectoraidb.actian.com/docs/fundamentals/collections/collections.md): Learn about collections, the primary unit of storage in VectorAI DB, including structure, distance metrics, index parameters, and lifecycle states.
- [Collection workflow](https://docs.vectoraidb.actian.com/docs/fundamentals/collections/complete-workflow.md): A complete walkthrough of creating, populating, querying, maintaining, and deleting a collection.
- [Create a collection](https://docs.vectoraidb.actian.com/docs/fundamentals/collections/create-collection-task.md): Create collections with default or custom index parameters.
- [Delete a collection](https://docs.vectoraidb.actian.com/docs/fundamentals/collections/delete-collection-task.md): Permanently remove or recreate collections.
- [Get collection info](https://docs.vectoraidb.actian.com/docs/fundamentals/collections/get-collection-info-task.md): Retrieve collection metadata, statistics, and state.
- [Inspect and maintain collections](https://docs.vectoraidb.actian.com/docs/fundamentals/collections/manage-collection-state-task.md): Monitor collection state, flush data, optimize storage, and rebuild indexes.
- [Update collection parameters](https://docs.vectoraidb.actian.com/docs/fundamentals/collections/update-collection-task.md): Modify collection configuration after creation.
- [Configure cosine similarity](https://docs.vectoraidb.actian.com/docs/fundamentals/distance-metrics/cosine-similarity-task.md): Create a collection that measures angular similarity between vectors.
- [Distance metrics overview](https://docs.vectoraidb.actian.com/docs/fundamentals/distance-metrics/distance-metrics.md): Measure similarity between vectors using cosine similarity, Euclidean distance, or dot product.
- [Configure dot product](https://docs.vectoraidb.actian.com/docs/fundamentals/distance-metrics/dot-product-task.md): Create a collection that measures alignment and magnitude efficiently.
- [Configure Euclidean distance](https://docs.vectoraidb.actian.com/docs/fundamentals/distance-metrics/euclidean-distance-task.md): Create a collection that measures straight-line distance between vectors.
- [Combine filter types](https://docs.vectoraidb.actian.com/docs/fundamentals/filtering/combined-filter-task.md): Build complex queries with must, should, and must-not conditions.
- [Filtering overview](https://docs.vectoraidb.actian.com/docs/fundamentals/filtering/filtering.md): Narrow vector search results using must, should, and must-not metadata conditions.
- [Filter with must conditions](https://docs.vectoraidb.actian.com/docs/fundamentals/filtering/must-filter-task.md): Apply AND logic to require all conditions match.
- [Filter with must-not conditions](https://docs.vectoraidb.actian.com/docs/fundamentals/filtering/must-not-filter-task.md): Exclude results where conditions are true.
- [Filter with should conditions](https://docs.vectoraidb.actian.com/docs/fundamentals/filtering/should-filter-task.md): Apply OR logic to match at least one condition.
- [Distribution-Based Score Fusion](https://docs.vectoraidb.actian.com/docs/fundamentals/hybrid-search/dbsf-fusion-task.md): Combine semantic and keyword search results using Distribution-Based Score Fusion.
- [Hybrid RAG retrieval](https://docs.vectoraidb.actian.com/docs/fundamentals/hybrid-search/hybrid-rag-task.md): Combine original and reformulated question embeddings for improved RAG context retrieval.
- [Overview](https://docs.vectoraidb.actian.com/docs/fundamentals/hybrid-search/hybrid-search.md): Hybrid search in Actian VectorAI DB combines results from multiple search queries using fusion algorithms like Reciprocal Rank Fusion and Distribution-Based Score Fusion to improve retrieval quality.
- [Multimodel fusion](https://docs.vectoraidb.actian.com/docs/fundamentals/hybrid-search/multi-model-fusion-task.md): Combine search results from multiple embedding models using fusion.
- [Performance benchmarking](https://docs.vectoraidb.actian.com/docs/fundamentals/hybrid-search/performance-benchmark-task.md): Compare execution time between single search and hybrid search approaches.
- [Query variation fusion](https://docs.vectoraidb.actian.com/docs/fundamentals/hybrid-search/query-variation-task.md): Improve search robustness by fusing results from query vector variations.
- [Reciprocal Rank Fusion](https://docs.vectoraidb.actian.com/docs/fundamentals/hybrid-search/rrf-fusion-task.md): Combine results from multiple dense vector searches using Reciprocal Rank Fusion.
- [Introduction](https://docs.vectoraidb.actian.com/docs/fundamentals/index.md): Learn the core fundamentals of Actian VectorAI DB.
- [Configure HNSW parameters](https://docs.vectoraidb.actian.com/docs/fundamentals/indexing/configure-hnsw-task.md): Set HNSW parameters when creating a collection to control the tradeoff between search speed, recall accuracy, and memory usage in VectorAI DB.
- [Indexing overview](https://docs.vectoraidb.actian.com/docs/fundamentals/indexing/indexing.md): Learn how VectorAI DB uses HNSW indexing for fast approximate nearest neighbor search, including tunable parameters and tradeoffs between speed, accuracy, and memory.
- [Create payloads](https://docs.vectoraidb.actian.com/docs/fundamentals/payload/create-payload-task.md): Add JSON metadata when inserting points into a collection.
- [Search with payload filters](https://docs.vectoraidb.actian.com/docs/fundamentals/payload/filter-payload-task.md): Apply must, should, and must-not filter conditions to payload fields during vector search in VectorAI DB to return results that match both similarity and metadata criteria.
- [Payloads overview](https://docs.vectoraidb.actian.com/docs/fundamentals/payload/payload.md): Payloads are JSON objects attached to points in VectorAI DB. They store metadata such as categories, prices, and timestamps that enable hybrid search combining vector similarity with metadata filtering.
- [Update payloads](https://docs.vectoraidb.actian.com/docs/fundamentals/payload/update-payload-task.md): Modify payload metadata for existing points.
- [Delete points](https://docs.vectoraidb.actian.com/docs/fundamentals/points/delete-points-task.md): Remove points from a collection by ID and compact to reclaim storage.
- [Insert points](https://docs.vectoraidb.actian.com/docs/fundamentals/points/insert-points-task.md): Add individual points or batches of points to a collection.
- [Points overview](https://docs.vectoraidb.actian.com/docs/fundamentals/points/points.md): Points are the fundamental data units in Actian VectorAI DB, each containing a unique ID, a vector embedding, and an optional JSON payload. Covers point structure, supported operations, and how points relate to collections.
- [Retrieve points](https://docs.vectoraidb.actian.com/docs/fundamentals/points/retrieve-points-task.md): Fetch points from a collection by ID or through pagination.
- [Update points](https://docs.vectoraidb.actian.com/docs/fundamentals/points/update-points-task.md): Modify existing points with new vector and payload data.
- [Basic similarity search](https://docs.vectoraidb.actian.com/docs/fundamentals/search/basic-search-task.md): Search for the most similar vectors to your query.
- [Search with filters](https://docs.vectoraidb.actian.com/docs/fundamentals/search/filtered-search-task.md): Combine vector similarity with metadata conditions.
- [Search overview](https://docs.vectoraidb.actian.com/docs/fundamentals/search/search.md): Vector search in VectorAI DB finds the most similar vectors to a query using distance metrics. Covers search parameters, result fields, score interpretation, and performance optimization strategies.
- [Search with payload](https://docs.vectoraidb.actian.com/docs/fundamentals/search/search-with-payload-task.md): Include metadata in search results.
- [Search with vectors](https://docs.vectoraidb.actian.com/docs/fundamentals/search/search-with-vectors-task.md): Include vector embeddings in search results.
- [Complete workflow](https://docs.vectoraidb.actian.com/docs/fundamentals/semantic-search/complete-workflow.md): End-to-end semantic search pipeline with embedding, indexing, and multiple search strategies.
- [Filtered semantic search](https://docs.vectoraidb.actian.com/docs/fundamentals/semantic-search/filtered-semantic-search-task.md): Combine vector similarity with keyword and range filters.
- [Multiconstraint search](https://docs.vectoraidb.actian.com/docs/fundamentals/semantic-search/multi-constraint-search-task.md): Combine multiple metadata conditions with vector similarity.
- [Pure semantic search](https://docs.vectoraidb.actian.com/docs/fundamentals/semantic-search/pure-semantic-search-task.md): Search for documents by meaning using vector similarity.
- [Score threshold search](https://docs.vectoraidb.actian.com/docs/fundamentals/semantic-search/score-threshold-search-task.md): Filter semantic search results by minimum similarity score.
- [Overview](https://docs.vectoraidb.actian.com/docs/fundamentals/semantic-search/semantic-search.md): Semantic search in Actian VectorAI DB finds documents by meaning rather than keywords. Covers embedding-based retrieval, payload filtering, score thresholds, and multiconstraint search patterns for RAG pipelines.
- [Store vectors](https://docs.vectoraidb.actian.com/docs/fundamentals/vectors/create-vectors-task.md): Store dense vectors with metadata in a collection.
- [Search with vectors](https://docs.vectoraidb.actian.com/docs/fundamentals/vectors/search-vectors-task.md): Search a collection in VectorAI DB using a query vector to retrieve the most semantically similar points, ranked by similarity score.
- [Vectors overview](https://docs.vectoraidb.actian.com/docs/fundamentals/vectors/vectors.md): Convert unstructured data into searchable numerical embeddings.
- [Error handling](https://docs.vectoraidb.actian.com/docs/guides/error-handling.md): Understand error codes and exception types returned by VectorAI DB and how to handle them in your application.
- [Local UI](https://docs.vectoraidb.actian.com/docs/guides/gui-interface.md): Manage and interact with VectorAI DB through the built-in web interface.
- [Troubleshooting](https://docs.vectoraidb.actian.com/docs/guides/troubleshooting.md): Identify and fix common operational problems with VectorAI DB using guided diagnostic steps.
- [Overview](https://docs.vectoraidb.actian.com/docs/integrations/index.md): Connect VectorAI DB with embedding providers and AI frameworks to build semantic search, RAG pipelines, and AI-powered applications.
- [LangChain](https://docs.vectoraidb.actian.com/docs/integrations/langchain.md): Use Actian VectorAI DB as a vector store in LangChain for building RAG pipelines, semantic search, and AI-powered applications.
- [LlamaIndex](https://docs.vectoraidb.actian.com/docs/integrations/llama-index.md): Use Actian VectorAI DB as a vector store in LlamaIndex for building RAG pipelines, semantic search, and AI-powered applications.
- [Overview](https://docs.vectoraidb.actian.com/home/getting-started/overview.md): High-performance vector database for AI similarity search.
- [Docker setup for Python](https://docs.vectoraidb.actian.com/home/installation/instructions.md): Get VectorAI DB running locally using Docker in just a few minutes.
- [Quickstart](https://docs.vectoraidb.actian.com/home/quickstart/quickstart.md): Guide to help you get started with VectorAI DB in under five minutes.
- [EULA](https://docs.vectoraidb.actian.com/home/support/eula.md): End-User License Agreement.
- [FAQ](https://docs.vectoraidb.actian.com/home/support/faq.md): Frequently asked questions about VectorAI DB.
- [Licensing](https://docs.vectoraidb.actian.com/home/support/license.md): Information on VectorAI DB licensing and other legal matters.
- [Support](https://docs.vectoraidb.actian.com/home/support/support.md): Resources and contact options for VectorAI DB community and licensed customers.
- [Installation](https://docs.vectoraidb.actian.com/sdks/javascript/installation.md): Install the JavaScript SDK with npm install @actian/vectorai-client.
- [JavaScript SDK Quickstart](https://docs.vectoraidb.actian.com/sdks/javascript/quickstart.md): Get started with VectorAI DB using the JavaScript SDK.
- [JavaScript SDK reference](https://docs.vectoraidb.actian.com/sdks/javascript/reference.md): Core JavaScript SDK client, namespaces, options, filters, auth, batching, and errors.
- [Installation](https://docs.vectoraidb.actian.com/sdks/python/installation.md): Install the Python SDK with pip install actian-vectorai-client.
- [Python SDK Quickstart](https://docs.vectoraidb.actian.com/sdks/python/quickstart.md): Get started with VectorAI DB using the Python SDK.
- [Python SDK reference](https://docs.vectoraidb.actian.com/sdks/python/reference.md): Core Python SDK clients, namespaces, configuration, filters, batching, and errors.

## OpenAPI Specs

- [search-api](https://docs.vectoraidb.actian.com/openapi_prepared/search-api.yaml)
- [points-api](https://docs.vectoraidb.actian.com/openapi_prepared/points-api.yaml)
- [grouped-search-api](https://docs.vectoraidb.actian.com/openapi_prepared/grouped-search-api.yaml)
- [filters-api](https://docs.vectoraidb.actian.com/openapi_prepared/filters-api.yaml)
- [collections-api](https://docs.vectoraidb.actian.com/openapi_prepared/collections-api.yaml)
- [authentication-api](https://docs.vectoraidb.actian.com/openapi_prepared/authentication-api.yaml)