# Pinecone Docs > Pinecone is the vector database for building accurate, performant AI applications at scale in production. The platform includes Pinecone Database (vector search and retrieval), Pinecone Assistant (document-grounded chat and RAG), and Pinecone Marketplace (no-code knowledge apps, public preview). The latest stable API version is 2026-04. SDKs are available for Python, Node.js, Java, Go, .NET, and Rust. ## Getting started - [Documentation overview](https://docs.pinecone.io/guides/get-started/overview): Landing page with quickstart paths for Database, Assistant, and Marketplace. - [Database quickstart](https://docs.pinecone.io/guides/get-started/quickstart): Create an index, upsert records, and run your first search query in minutes. - [Assistant quickstart](https://docs.pinecone.io/guides/assistant/quickstart/sdk-quickstart): Build a document-grounded assistant that answers questions about your data. - [Marketplace quickstart](https://docs.pinecone.io/guides/marketplace/quickstart): Deploy a no-code knowledge application from a template (public preview). - [Key concepts](https://docs.pinecone.io/guides/get-started/concepts): Core terminology: indexes, namespaces, vectors, records, metadata, embeddings, and sparse/dense representations. - [Database architecture](https://docs.pinecone.io/guides/get-started/database-architecture): How Pinecone's serverless architecture stores and retrieves vectors at scale. - [Using Pinecone with AI coding tools](https://docs.pinecone.io/guides/get-started/ai-coding-tools): Set up Pinecone in Claude Code, Cursor, Windsurf, Copilot, and other AI-assisted IDEs. - [Test at scale](https://docs.pinecone.io/guides/get-started/test-at-scale): Load-test and benchmark your Pinecone index before going to production. ## Pinecone Database - [Indexing overview](https://docs.pinecone.io/guides/index-data/indexing-overview): How data is organized in Pinecone: indexes, namespaces, dense vectors, sparse vectors, and metadata. - [Create an index](https://docs.pinecone.io/guides/index-data/create-an-index): Create serverless indexes with integrated or external embedding models. - [Data modeling](https://docs.pinecone.io/guides/index-data/data-modeling): Design your schema, choose dimensions, pick a metric, and plan namespaces. - [Upsert data](https://docs.pinecone.io/guides/index-data/upsert-data): Write records with vectors and metadata to an index, with or without integrated embedding. - [Import data](https://docs.pinecone.io/guides/index-data/import-data): Bulk-load data from cloud storage (S3, GCS, Azure Blob) into an index. - [Check data freshness](https://docs.pinecone.io/guides/index-data/check-data-freshness): Verify that upserted data is queryable and check replication status. - [Implement multitenancy](https://docs.pinecone.io/guides/index-data/implement-multitenancy): Use namespaces to isolate data per tenant in a single index. - [Dedicated read nodes](https://docs.pinecone.io/guides/index-data/dedicated-read-nodes): Scale read throughput independently from writes using dedicated read replicas. ## Search - [Search overview](https://docs.pinecone.io/guides/search/search-overview): Compare search modes: semantic, lexical, hybrid, and full-text. Choose the right approach for your use case. - [Full-text search](https://docs.pinecone.io/guides/search/full-text-search): Search typed JSON documents using BM25 scoring and Lucene query syntax (public preview, 2026-01.alpha API). - [Semantic search](https://docs.pinecone.io/guides/search/semantic-search): Find records most similar in meaning using dense vectors and nearest-neighbor retrieval. - [Lexical search](https://docs.pinecone.io/guides/search/lexical-search): Find records matching exact words or phrases using sparse vectors. - [Hybrid search](https://docs.pinecone.io/guides/search/hybrid-search): Combine semantic and lexical search with reranking for the best of both approaches. - [Filter by metadata](https://docs.pinecone.io/guides/search/filter-by-metadata): Narrow search results using metadata filters on fields like category, date, or source. - [Rerank results](https://docs.pinecone.io/guides/search/rerank-results): Improve search relevance by reranking initial results with a cross-encoder model. ## Optimize and operate - [Increase relevance](https://docs.pinecone.io/guides/optimize/increase-relevance): Tune embedding models, metadata filters, and reranking to improve result quality. - [Increase throughput](https://docs.pinecone.io/guides/optimize/increase-throughput): Batch operations, use gRPC, and configure read replicas for higher QPS. - [Decrease latency](https://docs.pinecone.io/guides/optimize/decrease-latency): Reduce query latency with region selection, caching, and index configuration. - [Target an index](https://docs.pinecone.io/guides/manage-data/target-an-index): Connect to a specific index by name or host URL using the SDK or API. - [Manage indexes](https://docs.pinecone.io/guides/manage-data/manage-indexes): List, describe, configure, and delete indexes. - [Manage namespaces](https://docs.pinecone.io/guides/manage-data/manage-namespaces): Create, list, and delete namespaces within an index. - [Backups overview](https://docs.pinecone.io/guides/manage-data/backups-overview): Back up and restore indexes for disaster recovery, migration, and testing. - [Back up an index](https://docs.pinecone.io/guides/manage-data/back-up-an-index): Create manual or scheduled backups of your index data. - [Restore an index](https://docs.pinecone.io/guides/manage-data/restore-an-index): Restore an index from a backup to the same or a different project. - [Update data](https://docs.pinecone.io/guides/manage-data/update-data): Update vector values or metadata for existing records. - [Delete data](https://docs.pinecone.io/guides/manage-data/delete-data): Delete records by ID, metadata filter, or namespace. - [Fetch data](https://docs.pinecone.io/guides/manage-data/fetch-data): Retrieve records by ID to inspect stored vectors and metadata. - [List record IDs](https://docs.pinecone.io/guides/manage-data/list-record-ids): Paginate through all record IDs in a namespace. - [Understanding cost](https://docs.pinecone.io/guides/manage-cost/understanding-cost): How Pinecone pricing works: read units, write units, storage, and compute. - [Manage cost](https://docs.pinecone.io/guides/manage-cost/manage-cost): Strategies to control spend: right-size indexes, use namespaces, monitor usage. - [Monitor usage and costs](https://docs.pinecone.io/guides/manage-cost/monitor-usage-and-costs): Track usage and spending in the Pinecone console and via API. ## Production - [Production checklist](https://docs.pinecone.io/guides/production/production-checklist): Pre-launch checklist covering security, scaling, monitoring, and error handling. - [Bring your own cloud (BYOC)](https://docs.pinecone.io/guides/production/bring-your-own-cloud): Run Pinecone in your own AWS or Azure account for data residency and compliance (public preview). - [Security overview](https://docs.pinecone.io/guides/production/security-overview): Authentication, encryption, SSO, CMEK, private endpoints, audit logs, and compliance. - [Configure SSO with Okta](https://docs.pinecone.io/guides/production/configure-single-sign-on/okta): Set up SAML-based single sign-on with Okta for your Pinecone organization. - [Configure CMEK](https://docs.pinecone.io/guides/production/configure-cmek): Encrypt index data with your own AWS KMS key. - [Configure private endpoints](https://docs.pinecone.io/guides/production/configure-private-endpoints): Connect to Pinecone over AWS PrivateLink or Azure Private Link for network isolation. - [Data deletion](https://docs.pinecone.io/guides/production/data-deletion): How Pinecone handles data deletion, retention, and purging. - [Configure audit logs](https://docs.pinecone.io/guides/production/configure-audit-logs): Stream organization activity to your SIEM or log management platform. - [Error handling](https://docs.pinecone.io/guides/production/error-handling): Handle rate limits, timeouts, and transient errors with retries and backoff. - [Monitoring](https://docs.pinecone.io/guides/production/monitoring): Monitor index health, latency, and throughput with Prometheus metrics and the console dashboard. - [Automated testing](https://docs.pinecone.io/guides/production/automated-testing): Test Pinecone integrations in CI/CD pipelines. ## Pinecone Assistant - [Assistant overview](https://docs.pinecone.io/guides/assistant/overview): Build production-grade chat and agent-based applications grounded in your documents. - [Create an assistant](https://docs.pinecone.io/guides/assistant/create-assistant): Create and configure assistants with custom instructions and model selection. - [Manage assistants](https://docs.pinecone.io/guides/assistant/manage-assistants): List, update, and delete assistants. - [Files overview](https://docs.pinecone.io/guides/assistant/files-overview): Supported file types, size limits, and how files are processed. - [Upload files](https://docs.pinecone.io/guides/assistant/upload-files): Upload local files, URLs, or cloud storage objects to an assistant's knowledge base. - [Multimodal files](https://docs.pinecone.io/guides/assistant/multimodal): Upload and process images, charts, and mixed-content documents. - [Manage files](https://docs.pinecone.io/guides/assistant/manage-files): List, describe, and delete files attached to an assistant. - [Chat with an assistant](https://docs.pinecone.io/guides/assistant/chat-with-assistant): Send messages and receive grounded answers with inline citations. - [OpenAI-compatible chat](https://docs.pinecone.io/guides/assistant/chat-through-the-openai-compatible-interface): Use the OpenAI chat completions format to chat with a Pinecone Assistant. - [Evaluation overview](https://docs.pinecone.io/guides/assistant/evaluation-overview): Measure answer quality with alignment and completeness metrics. - [Evaluate answers](https://docs.pinecone.io/guides/assistant/evaluate-answers): Run evaluation metrics against assistant responses. - [Context snippets overview](https://docs.pinecone.io/guides/assistant/context-snippets-overview): Retrieve relevant document chunks without generating a full answer, for use in custom RAG pipelines. - [Retrieve context snippets](https://docs.pinecone.io/guides/assistant/retrieve-context-snippets): Call the context endpoint to get ranked snippets from your assistant's files. - [Pricing and limits](https://docs.pinecone.io/guides/assistant/pricing-and-limits): Assistant pricing tiers, rate limits, file size limits, and usage quotas. - [Assistant architecture](https://docs.pinecone.io/reference/architecture/assistant-architecture): How the Assistant service processes files, generates embeddings, and retrieves context. ## Pinecone Marketplace (public preview) - [Marketplace overview](https://docs.pinecone.io/guides/marketplace/overview): No-code platform for creating, publishing, and operating knowledge applications. - [Marketplace concepts](https://docs.pinecone.io/guides/marketplace/concepts): Key concepts: deployments, templates, connectors, versions, and domains. - [Quickstart](https://docs.pinecone.io/guides/marketplace/quickstart): Deploy your first knowledge application from a template. - [Pricing and limits](https://docs.pinecone.io/guides/marketplace/pricing-and-limits): Marketplace pricing, deployment limits, and connector quotas. - [Templates overview](https://docs.pinecone.io/guides/marketplace/templates-overview): Browse and configure pre-built templates for common knowledge app patterns. - [Template catalog](https://docs.pinecone.io/guides/marketplace/template-catalog): Full catalog of available templates with descriptions and use cases. - [Create a deployment](https://docs.pinecone.io/guides/marketplace/create-a-deployment): Configure and launch a knowledge application from a template. - [Configure operating parameters](https://docs.pinecone.io/guides/marketplace/configure-operating-parameters): Set LLM, retrieval, and response parameters for a deployment. - [Configure layouts](https://docs.pinecone.io/guides/marketplace/configure-layouts): Customize the look and structure of your knowledge application. - [Configure components](https://docs.pinecone.io/guides/marketplace/configure-components): Add and arrange UI components like search bars, chat widgets, and result cards. - [Connectors overview](https://docs.pinecone.io/guides/marketplace/connectors-overview): Connect external data sources (web crawlers, file stores, APIs) to a deployment. - [Multi-domain routing](https://docs.pinecone.io/guides/marketplace/multi-domain-routing): Route queries across multiple knowledge domains with automatic domain detection. - [Publish a deployment](https://docs.pinecone.io/guides/marketplace/publish-a-deployment): Make your knowledge application available to end users via a public URL. - [Manage versions and rollback](https://docs.pinecone.io/guides/marketplace/manage-versions-and-rollback): Version your deployment configuration and roll back to a previous version. - [Evaluations](https://docs.pinecone.io/guides/marketplace/evaluations): Measure answer quality and retrieval accuracy for a deployment. - [Analytics and event logs](https://docs.pinecone.io/guides/marketplace/analytics-and-event-logs): Track usage, queries, and events for a deployment. - [Deployer auth](https://docs.pinecone.io/guides/marketplace/deployer-auth): Configure authentication for deployment administrators. - [Consumer auth](https://docs.pinecone.io/guides/marketplace/consumer-auth-overview): Configure authentication for end users of your knowledge application. ## AI agent integration - [Pinecone MCP server (Database)](https://docs.pinecone.io/guides/operations/mcp-server): Connect AI agents and IDEs to Pinecone Database via the Model Context Protocol. - [Assistant MCP server](https://docs.pinecone.io/guides/assistant/mcp-server): Connect AI agents to Pinecone Assistant via the Model Context Protocol. - [Agent skills](https://docs.pinecone.io/integrations/agent-skills): Use Pinecone as a tool in agentic workflows with pre-built skill definitions. - [Claude Code](https://docs.pinecone.io/integrations/claude-code): Use Pinecone from Claude Code with the MCP integration. - [Gemini CLI](https://docs.pinecone.io/integrations/gemini-cli): Use Pinecone from the Gemini CLI. - [GitHub Copilot](https://docs.pinecone.io/integrations/github-copilot): Use Pinecone with GitHub Copilot extensions. ## SDKs and CLI - [SDK overview](https://docs.pinecone.io/reference/pinecone-sdks): Supported languages, installation, and version compatibility. - [Python SDK](https://docs.pinecone.io/reference/sdks/python/overview): Install, configure, and use the Python client. Supports gRPC for high-throughput workloads. - [Node.js SDK](https://docs.pinecone.io/reference/sdks/node/overview): Install, configure, and use the TypeScript/JavaScript client. - [Java SDK](https://docs.pinecone.io/reference/sdks/java/overview): Install, configure, and use the Java client. Includes OpenTelemetry support. - [Go SDK](https://docs.pinecone.io/reference/sdks/go/overview): Install, configure, and use the Go client. - [.NET SDK](https://docs.pinecone.io/reference/sdks/dotnet/overview): Install, configure, and use the C#/.NET client. - [Rust SDK](https://docs.pinecone.io/reference/sdks/rust/overview): Install, configure, and use the Rust client. - [CLI quickstart](https://docs.pinecone.io/reference/cli/quickstart): Install the Pinecone CLI and run your first commands. - [CLI command reference](https://docs.pinecone.io/reference/cli/command-reference): Full list of CLI commands with usage and flags. ## API reference (2026-04, latest stable) - [API introduction](https://docs.pinecone.io/reference/api/introduction): Base URLs, request format, response format, and pagination. - [Authentication](https://docs.pinecone.io/reference/api/authentication): API key authentication and OAuth2 service account tokens. - [API versioning](https://docs.pinecone.io/reference/api/versioning): How Pinecone versions its API, version lifecycle, and deprecation policy. - [Database limits](https://docs.pinecone.io/reference/api/database-limits): Max dimensions, metadata size, batch sizes, and rate limits. - [Known limitations](https://docs.pinecone.io/reference/api/known-limitations): Current platform limitations and workarounds. - [Errors](https://docs.pinecone.io/reference/api/errors): Error codes, messages, and troubleshooting guidance. - [Assistant API introduction](https://docs.pinecone.io/reference/api/assistant/introduction): Base URLs and request format for the Assistant API. - [Assistant authentication](https://docs.pinecone.io/reference/api/assistant/authentication): API key authentication for the Assistant API. - [Assistant limits](https://docs.pinecone.io/reference/api/assistant/assistant-limits): Rate limits, file size limits, and quotas for the Assistant API. - [Marketplace API introduction](https://docs.pinecone.io/reference/api/marketplace/introduction): Base URLs and request format for the Marketplace API (public preview). Older API versions (2025-10, 2025-04, 2025-01, 2024-10, 2024-07, 2024-04) are available at `https://docs.pinecone.io/reference/api/{version}/...` ## Inference (hosted models) - [Model gallery](https://docs.pinecone.io/models/overview): Browse all embedding and reranking models hosted by Pinecone, with specs and usage. - [Generate embeddings API](https://docs.pinecone.io/reference/api/2026-04/inference/generate-embeddings): Convert text to dense or sparse vectors using Pinecone-hosted embedding models. - [Rerank API](https://docs.pinecone.io/reference/api/2026-04/inference/rerank): Rerank a list of documents by relevance to a query using a hosted cross-encoder model. - [List models API](https://docs.pinecone.io/reference/api/2026-04/inference/list_models): List all available inference models. - [Describe model API](https://docs.pinecone.io/reference/api/2026-04/inference/describe_model): Get details about a specific inference model. ## Integrations - [Integrations overview](https://docs.pinecone.io/integrations/overview): Browse all partner integrations across data sources, frameworks, infrastructure, models, and observability. - [LangChain](https://docs.pinecone.io/integrations/langchain): Use Pinecone as a vector store in LangChain for RAG and agent workflows. - [LlamaIndex](https://docs.pinecone.io/integrations/llamaindex): Use Pinecone as a vector store in LlamaIndex data pipelines. - [Haystack](https://docs.pinecone.io/integrations/haystack): Use Pinecone as a document store in Haystack NLP pipelines. - [Amazon Bedrock](https://docs.pinecone.io/integrations/amazon-bedrock): Connect Pinecone as a knowledge base for Amazon Bedrock agents. - [OpenAI](https://docs.pinecone.io/integrations/openai): Use OpenAI embeddings with Pinecone for semantic search and RAG. - [Databricks](https://docs.pinecone.io/integrations/databricks): Ingest data from Databricks into Pinecone. - [Snowflake](https://docs.pinecone.io/integrations/snowflake): Ingest data from Snowflake into Pinecone. - [Terraform](https://docs.pinecone.io/integrations/terraform): Manage Pinecone resources as infrastructure-as-code with the Terraform provider. - [Pulumi](https://docs.pinecone.io/integrations/pulumi): Manage Pinecone resources with the Pulumi provider. - [Vercel](https://docs.pinecone.io/integrations/vercel): Deploy Pinecone-powered apps on Vercel with the AI SDK integration. - [n8n](https://docs.pinecone.io/integrations/n8n): Build Pinecone workflows visually with the n8n automation platform. - [Datadog](https://docs.pinecone.io/integrations/datadog): Monitor Pinecone metrics and logs in Datadog. - [Build an integration](https://docs.pinecone.io/integrations/build-integration/integration-ecosystem): Build and publish your own Pinecone integration. ## Examples - [Notebooks](https://docs.pinecone.io/examples/notebooks): Colab notebooks for search, RAG, embedding, reranking, data loading, and more. - [Sample apps](https://docs.pinecone.io/examples/sample-apps): Full-stack sample applications: semantic search, multi-tenant RAG, multimodal search, and Assistant chat. - [Reference architectures](https://docs.pinecone.io/examples/reference-architectures): AWS reference architecture for high-scale production systems using Pinecone. - [Assistant examples](https://docs.pinecone.io/examples/assistant): Notebooks and sample apps for Pinecone Assistant: quickstart, context snippets, and chat UI. ## Admin - [Understanding organizations](https://docs.pinecone.io/guides/organizations/understanding-organizations): How organizations, projects, members, and roles work in Pinecone. - [Manage API keys](https://docs.pinecone.io/guides/projects/manage-api-keys): Create, rotate, and delete API keys for your project. - [Manage service accounts](https://docs.pinecone.io/guides/organizations/manage-service-accounts): Create service accounts for programmatic access with OAuth2 tokens. - [Understanding projects](https://docs.pinecone.io/guides/projects/understanding-projects): How projects organize indexes, API keys, and billing within an organization. - [Upgrade billing plan](https://docs.pinecone.io/guides/organizations/manage-billing/upgrade-billing-plan): Upgrade from the free Starter plan to a paid plan. - [Standard trial](https://docs.pinecone.io/guides/organizations/manage-billing/standard-trial): How the free trial works: credits, limits, and what happens when it expires. ## Troubleshooting - [Contact support](https://docs.pinecone.io/troubleshooting/contact-support): How to reach Pinecone support via the console, email, or Slack. - [How to work with support](https://docs.pinecone.io/troubleshooting/how-to-work-with-support): What information to include when filing a support ticket for faster resolution. - [Support SLAs](https://docs.pinecone.io/troubleshooting/pinecone-support-slas): Response time targets by plan tier and severity level. ## Optional - [Full llms.txt index (auto-generated)](https://docs.pinecone.io/llms.txt): Complete machine-readable page list with titles, URLs, and descriptions. Auto-generated by Mintlify from all docs pages. - [Release notes (Database, 2026)](https://docs.pinecone.io/release-notes/2026): Latest Database release notes. - [Release notes (Assistant, 2026)](https://docs.pinecone.io/assistant-release-notes/2026): Latest Assistant release notes. - [Feature availability](https://docs.pinecone.io/release-notes/feature-availability): Which features are GA, public preview, or early access. - [Using pods (legacy)](https://docs.pinecone.io/guides/indexes/pods/understanding-pod-based-indexes): Pod-based indexes are the legacy deployment model. Serverless is now the default. - [Migrate pods to serverless](https://docs.pinecone.io/guides/indexes/pods/migrate-a-pod-based-index-to-serverless): Step-by-step guide to migrating from pod-based to serverless indexes. - [Cloud storage integrations](https://docs.pinecone.io/guides/operations/integrations/integrate-with-amazon-s3): Connect Pinecone to S3, GCS, or Azure Blob Storage for bulk data import. - [Local development](https://docs.pinecone.io/guides/operations/local-development): Run Pinecone locally for development and testing with Pinecone Local. - [Spark connector](https://docs.pinecone.io/reference/tools/pinecone-spark-connector): Load data from Apache Spark into Pinecone. - [2026-01.alpha API (full-text search)](https://docs.pinecone.io/reference/api/2026-01.alpha/control-plane/list_indexes): Public preview API version for full-text search with typed document schemas.