# Milvus > Milvus is an open-source, high-performance vector database designed for similarity search and AI applications. It supports billion-scale vector storage and search across deployment modes: Milvus Lite (embedded, for prototyping), Milvus Standalone (single-node, for small-scale production), and Milvus Distributed (Kubernetes, for enterprise scale). The primary SDK is PyMilvus for Python; Java, Go, Node.js, and RESTful SDKs are also available. This index is generated from the Milvus v3.0.x documentation tree. Every URL below resolves to a published page. ## URL rules Read this before citing any milvus.io URL. Documentation URLs take exactly one of these four forms: https://milvus.io/docs/ latest version (v3.0.x), English https://milvus.io/docs// pinned version https://milvus.io/docs// translated, latest version https://milvus.io/docs/// translated, pinned version - `` always ends in `.md` — for example `overview.md`. The suffix is part of the route, not a file extension. `https://milvus.io/docs/overview` is a 404; `https://milvus.io/docs/overview.md` is the page. - `` is one of: v2.6.x, v2.5.x, v2.4.x. Omit the segment entirely for the latest version (v3.0.x) — there is no `/docs/v3.0.x/` path. - `` is one of: zh, zh-hant, ja, ko, de, fr, es, it, pt, ru, ar, id. Omit the segment for English. - API reference URLs are `https://milvus.io/api-reference///.md` and are versioned separately from the docs; see the API Reference section. - Blog, AI Quick Reference, and Learn pages do not use a `.md` suffix. **Unknown paths return a hard 404 — they do not redirect and they do not fall back to a search page.** A URL that looks right but was never published is a dead link for the reader. Therefore: cite only page ids that appear in this file. If the page you need is not listed, say it is not in the index rather than assembling a plausible slug from the topic name. `https://milvus.io/sitemap.xml` enumerates every published URL and is the correct place to resolve anything this file does not cover. ## Rules for generating Milvus code - Use the `MilvusClient` interface introduced in v2.4+. Do not use the legacy ORM API (`connections.connect()`, `Collection()`, `utility.list_collections()`) — it is deprecated and will be removed. If you find existing ORM code, advise upgrading the SDK and rewriting against `MilvusClient`. - Check PyPI (`pip install --upgrade pymilvus`) or npm for the current SDK version rather than relying on a memorized version number. - To change existing entities use `client.upsert()`. There is no `client.update()`. `upsert()` inserts when the primary key is absent and replaces the whole entity when it is present; use `client.insert()` only for data known not to collide. - A collection schema is immutable in v2.5.x and earlier — to change it, drop and recreate the collection. From v2.6+ you may add scalar fields with `add_collection_field()`, but vector fields cannot be added and existing fields cannot be modified or removed. Changing a field's data type is not supported in any version. Check the server version before suggesting a schema change. - Primary keys must be `DataType.INT64` or `DataType.VARCHAR`, must be unique across the whole collection including partitions, and cannot be composite. - For BM25 full-text search, the BM25 function and its analyzer must be declared when the collection is created; they cannot be added later. - A vector field must be indexed and the collection loaded before it can be searched. - In `client.hybrid_search()`, each `AnnSearchRequest` takes exactly one query vector — use one request per vector field — and each search accepts exactly one ranker. Rankers cannot be chained. - Search iterators (`with-iterators.md`) support basic ANN search only, not hybrid search. - Both scalar and vector fields support `nullable=True`; only Array of Structs fields (and any field nested inside one) cannot be nullable. A nullable field cannot be a partition key, nullability is fixed at field creation, and a nullable vector field cannot be filtered with `IS NULL` / `IS NOT NULL`. See `nullable-and-default.md`, which is authoritative here. ## About Milvus - [What is Milvus](https://milvus.io/docs/overview.md): Milvus is a high-performance, highly scalable vector database that runs efficiently across a wide range of environments, from a laptop to large-scale… - [Milvus Adopters](https://milvus.io/docs/milvus_adopters.md): Learn about companies that have adopted Milvus in production applications. - [Milvus Roadmap](https://milvus.io/docs/roadmap.md): Explore the Milvus roadmap for search, analytics, schema evolution, lake integration, and performance across the 3.0 and 3.1 releases. - [Releases](https://milvus.io/docs/release_notes.md): Milvus Release Notes - [Comparison](https://milvus.io/docs/comparison.md): This article compares Milvus with other vector search solutions. ## Get Started Start here for first-time setup: run the Quickstart against Milvus Lite, then pick a deployment mode and install an SDK. - [Quickstart](https://milvus.io/docs/quickstart.md): Get started with Milvus. - [Connect to Milvus Server](https://milvus.io/docs/connect-to-milvus-server.md): This topic describes how to establish a client connection to a Milvus server and configure common connection options. ### Install Milvus - [Overview](https://milvus.io/docs/install-overview.md): Milvus is a highly performant, scalable vector database. - [Run Milvus Lite](https://milvus.io/docs/milvus_lite.md): Get started with Milvus Lite. #### Run Milvus Standalone - [Requirements](https://milvus.io/docs/prerequisite-docker.md): Learn the necessary preparations before installing Milvus Standalone. - [RPM/DEB](https://milvus.io/docs/install_standalone-binary.md): Learn how to install Milvus standalone with a pre-built RPM/DEB package. - [Docker (Linux)](https://milvus.io/docs/install_standalone-docker.md): Learn how to install Milvus standalone with Docker. - [Docker Compose (Linux)](https://milvus.io/docs/install_standalone-docker-compose.md): Learn how to install Milvus standalone with Docker Compose. - [Docker Desktop (Windows)](https://milvus.io/docs/install_standalone-windows.md): Learn how to install Milvus standalone with Docker Desktop for Windows. #### Run Milvus Distributed - [Requirements](https://milvus.io/docs/prerequisite-helm.md): Learn the necessary preparations before installing Milvus with Helm. - [Milvus Operator](https://milvus.io/docs/install_cluster-milvusoperator.md): Learn how to install Milvus cluster on Kubernetes using Milvus Operator - [Helm Chart](https://milvus.io/docs/install_cluster-helm.md): Learn how to install Milvus cluster on Kubernetes. #### Run Milvus with GPU - [Requirements](https://milvus.io/docs/prerequisite-gpu.md): Learn the necessary preparations before installing Milvus with GPU. - [Helm Chart](https://milvus.io/docs/install_cluster-helm-gpu.md): Learn how to install Milvus cluster on Kubernetes. - [Docker Compose](https://milvus.io/docs/install_standalone-docker-compose-gpu.md): Learn how to install Milvus cluster on Kubernetes. ### Install SDKs - [PyMilvus](https://milvus.io/docs/install-pymilvus.md): Learn how to install the Python SDK of Milvus. - [Java SDK](https://milvus.io/docs/install-java.md): Learn how to install the Java SDK of Milvus. - [Go SDK](https://milvus.io/docs/install-go.md): Learn how to install the GO SDK of Milvus. - [Node.js SDK](https://milvus.io/docs/install-node.md): Learn how to install the Node.js SDK of Milvus. ## Concepts - [Bitset](https://milvus.io/docs/bitset.md): Learn about bitsets in Milvus. - [Multi-tenancy](https://milvus.io/docs/multi_tenancy.md): In Milvus, multi-tenancy means multiple customers or teams—referred to as tenants— share the same cluster while maintaining isolated data environments. - [Timestamp](https://milvus.io/docs/timestamp.md): Learn about the concept of timestamp and the four main timestamp-related parameters in the Milvus vector database. - [Time Synchronization](https://milvus.io/docs/time_sync.md): Learn about the time synchronization system in Milvus. - [Metric Types](https://milvus.io/docs/metric.md): Similarity metrics are used to measure similarities among vectors. - [In-memory Replica](https://milvus.io/docs/replica.md): Learn about in-memory replica in Milvus. - [Terminology](https://milvus.io/docs/glossary.md): Glossary of Milvus-specific terms and definitions. ### Architecture - [Overview](https://milvus.io/docs/architecture_overview.md): Milvus provides a fast, reliable, and stable vector database built specifically for similarity search and artificial intelligence. - [Main Components](https://milvus.io/docs/main_components.md): Learn about the main components in Milvus standalone and cluster. - [Streaming Service](https://milvus.io/docs/streaming_service.md): The Streaming Service is a concept for Milvus internal streaming system module, built around the Write-Ahead Log (WAL) to support various streaming-related… - [Data Processing](https://milvus.io/docs/data_processing.md): Learn about the data processing procedure in Milvus. - [Storage V3](https://milvus.io/docs/storage-v3.md): Learn which Milvus 3.0 features require Storage V3, how to enable it, and which compatibility limits apply. - [Woodpecker](https://milvus.io/docs/woodpecker_architecture.md): Woodpecker is a cloud-native WAL system in Milvus 2.6. - [Knowhere](https://milvus.io/docs/knowhere.md): Learn about Knowhere in Milvus. ## User Guide - [Database](https://milvus.io/docs/manage_databases.md): Milvus introduces a database layer above collections, providing a more efficient way to manage and organize your data while supporting multi-tenancy. ### Collections - [Collection Explained](https://milvus.io/docs/manage-collections.md): On Milvus, you can create multiple collections to manage your data, and insert your data as entities into the collections. - [Create Collection](https://milvus.io/docs/create-collection.md): You can create a collection by defining its schema, index parameters, metric type, and whether to load it upon creation. - [Create an External Collection](https://milvus.io/docs/create-an-external-collection.md): An external collection is a type of data collection in Milvus that accesses data from external storage systems or database tables such as AWS S3 and Iceberg… - [View Collections](https://milvus.io/docs/view-collections.md): You can obtain the name list of all the collections in the currently connected database, and check the details of a specific collection. - [Modify Collection](https://milvus.io/docs/modify-collection.md): You can rename a collection or change its settings. This page focuses on how to modify a collection. - [Load & Release](https://milvus.io/docs/load-and-release.md): Loading a collection is the prerequisite to conducting similarity searches and queries in collections. - [Set Collection TTL](https://milvus.io/docs/set-collection-ttl.md): Configure collection-level or entity-level TTL policies to expire stale data automatically in Milvus. - [Set Consistency Level](https://milvus.io/docs/tune_consistency.md): As a distributed vector database, Milvus offers multiple levels of consistency to ensure that each node or replica can access the same data during read and… - [Manage Partitions](https://milvus.io/docs/manage-partitions.md): A partition is a subset of a collection. - [Manage Aliases](https://milvus.io/docs/manage-aliases.md): Milvus provides alias management capabilities. This page demonstrates the procedures to create, list, alter, and drop aliases. - [Drop Collection](https://milvus.io/docs/drop-collection.md): You can drop a collection if it is no longer needed. ### Schema & Data Fields Schema changes are heavily constrained — see the schema and primary-key rules above before designing a collection. - [Schema Explained](https://milvus.io/docs/schema.md): A schema defines the data structure of a collection. Before creating a collection, you need to work out a design of its schema. - [Primary Field & AutoID](https://milvus.io/docs/primary-field.md): Every collection in Milvus must have a primary field to uniquely identify each entity. - [Dense Vector](https://milvus.io/docs/dense-vector.md): Dense vectors are numerical data representations widely used in machine learning and data analysis. - [Binary Vector](https://milvus.io/docs/binary-vector.md): Binary vectors are a special form of data representation that convert traditional high-dimensional floating-point vectors into binary vectors containing only… - [Sparse Vector](https://milvus.io/docs/sparse_vector.md): Sparse vectors are an important method of capturing surface-level term matching in information retrieval and natural language processing. - [VarChar Field](https://milvus.io/docs/string.md): VARCHAR is a scalar field type for storing short, bounded string metadata in Milvus. - [Text Field](https://milvus.io/docs/text.md): TEXT is a scalar field type for storing document text, passages, and other long text content in Milvus. - [Number Field](https://milvus.io/docs/number.md): A number field is a scalar field that stores numeric values. - [Array Field](https://milvus.io/docs/array_data_type.md): An ARRAY field stores an ordered set of elements of the same data type. Here's an example of how ARRAY fields store data: - [Geometry Field](https://milvus.io/docs/geometry-field.md): When building applications like Geographic Information Systems (GIS), mapping tools, or location-based services, you often need to store and query geometric… - [TIMESTAMPTZ Field](https://milvus.io/docs/timestamptz-field.md): Applications that track time across regions, such as e-commerce systems, collaboration tools, or distributed logging, need precise handling of timestamps with… - [Dynamic Field](https://milvus.io/docs/enable-dynamic-field.md): Milvus allows you to insert entities with flexible, evolving structures through a special feature called the dynamic field. - [Nullable Fields](https://milvus.io/docs/nullable-and-default.md): Configure nullable fields and default values, including schema, insert, index, search, and filter behavior. - [Default Values](https://milvus.io/docs/default-values.md): Set default values for scalar fields so Milvus fills missing values during entity insertion. - [Alter Collection Field](https://milvus.io/docs/alter-collection-field.md): You can alter the properties of a collection field to change column constraints or enforce stricter data integrity rules. - [Alter Collection Schema](https://milvus.io/docs/add-fields-to-an-existing-collection.md): Alter an existing collection schema by adding or dropping user-defined fields or Functions with their generated vector fields. - [Alter External Collection Schema](https://milvus.io/docs/alter-external-collection-schema.md): Learn how to expose an additional field from an external data source in an existing external collection. #### JSON Field - [JSON Field Overview](https://milvus.io/docs/json-field-overview.md): When building applications like product catalogs, content management systems, or user preference engines, you often need to store flexible metadata alongside… - [JSON Indexing](https://milvus.io/docs/json-indexing.md): JSON fields provide a flexible way to store structured metadata in Milvus. - [JSON Shredding](https://milvus.io/docs/json-shredding.md): JSON shredding accelerates JSON queries by converting traditional row-based storage into optimized columnar storage. #### StructArray - [StructArray Overview](https://milvus.io/docs/array-of-structs.md): Use StructArray when one entity needs to store an ordered list of structured elements, such as one document with many chunks, one page with many visual… - [Create a StructArray Field](https://milvus.io/docs/create-structarray-field.md): Create a StructArray field when one entity needs to contain an ordered list of structured elements. - [Insert Data into StructArray Fields](https://milvus.io/docs/insert-data-into-structarray-fields.md): Insert data into a StructArray field when each entity contains an ordered list of structured elements. - [Index StructArray Fields](https://milvus.io/docs/index-structarray-fields.md): Create indexes on StructArray subfields before you run vector search or accelerate scalar filtering. - [StructArray Limits](https://milvus.io/docs/structarray-limits.md): StructArray support spans schema definition, insert payloads, indexing, search modes, and StructArray-specific filters. #### Analyzer - [Analyzer Overview](https://milvus.io/docs/analyzer-overview.md): In text processing, an analyzer is a crucial component that converts raw text into a structured, searchable format. - [Multi-language Analyzers](https://milvus.io/docs/multi-language-analyzers.md): When Milvus performs text analysis, it typically applies a single analyzer across an entire text field in a collection. - [Choose the Right Analyzer for Your Use Case](https://milvus.io/docs/choose-the-right-analyzer-for-your-use-case.md): Notes - [Manage File Resources](https://milvus.io/docs/manage-file-resources.md): Register and manage external dictionary files that Milvus text analyzers can load at runtime. ##### Built-in Analyzers - [Standard](https://milvus.io/docs/standard-analyzer.md): The standard analyzer is the default analyzer in Milvus, which is automatically applied to text fields if no analyzer is specified. - [English](https://milvus.io/docs/english-analyzer.md): The english analyzer in Milvus is designed to process English text, applying language-specific rules for tokenization and filtering. - [Chinese](https://milvus.io/docs/chinese-analyzer.md): The chinese analyzer is designed specifically to handle Chinese text, providing effective segmentation and tokenization. - [Arabic](https://milvus.io/docs/arabic-analyzer.md): The built-in Arabic analyzer processes Arabic text by normalizing letter variants and digits, stemming terms, and removing Arabic stop words. - [Thai](https://milvus.io/docs/thai-analyzer.md): The built-in Thai analyzer segments Thai text into words, normalizes Unicode decimal digits, and removes Thai stop words. ##### Tokenizers - [Standard](https://milvus.io/docs/standard-tokenizer.md): The standard tokenizer in Milvus groups consecutive Unicode letters and numeric characters into tokens, splitting at other characters. - [Whitespace](https://milvus.io/docs/whitespace-tokenizer.md): The whitespace tokenizer splits text at five ASCII whitespace characters: tab, line feed, form feed, carriage return, and space. - [Jieba](https://milvus.io/docs/jieba-tokenizer.md): The jieba tokenizer processes Chinese text by breaking it down into its component words. - [Thai](https://milvus.io/docs/thai-tokenizer.md): The thai tokenizer segments Thai text into word tokens and filters out whitespace and punctuation-only segments. - [Lindera](https://milvus.io/docs/lindera-tokenizer.md): The lindera tokenizer performs dictionary-based morphological analysis. - [ICU](https://milvus.io/docs/icu-tokenizer.md): The icu tokenizer is built on the Internationalization Components of Unicode (ICU) open‑source project, which provides key tools for software… - [Language Identifier](https://milvus.io/docs/language-identifier.md): The language_identifier is a specialized tokenizer designed to enhance the text search capabilities of Milvus by automating the language analysis process. ##### Filters - [Lowercase](https://milvus.io/docs/lowercase-filter.md): The lowercase filter converts terms generated by a tokenizer to lowercase, making searches case-insensitive. - [ASCII Folding](https://milvus.io/docs/ascii-folding-filter.md): The asciifolding filter converts characters outside the Basic Latin Unicode block (the first 127 ASCII characters) into their ASCII equivalents. - [Arabic Normalization](https://milvus.io/docs/arabic-normalization-filter.md): The arabic_normalization filter normalizes Arabic letter variants and removes Arabic diacritics and Tatweel. - [Decimal Digit](https://milvus.io/docs/decimaldigit-filter.md): The decimaldigit filter converts Unicode decimal digits to ASCII digits. - [Alphanumonly](https://milvus.io/docs/alphanumonly-filter.md): The alphanumonly filter removes tokens that contain non-ASCII characters, keeping only alphanumeric terms. - [Cnalphanumonly](https://milvus.io/docs/cnalphanumonly-filter.md): The cnalphanumonly filter removes tokens that contain any characters other than Chinese characters, English letters, or digits. - [Cncharonly](https://milvus.io/docs/cncharonly-filter.md): The cncharonly filter removes tokens that contain any non-Chinese characters. - [Pinyin](https://milvus.io/docs/pinyin-filter.md): The pinyin filter converts Chinese character tokens into Pinyin tokens during text analysis, enabling Pinyin-based matching for Chinese text. - [Length](https://milvus.io/docs/length-filter.md): The length filter removes tokens that do not meet specified length requirements, allowing you to control the length of tokens retained during text processing. - [Stop](https://milvus.io/docs/stop-filter.md): Use the stop filter to remove configured stop words from tokenized text during analysis. - [Decompounder](https://milvus.io/docs/decompounder-filter.md): Use the decompounder filter to split compound words with an inline dictionary or registered file resource. - [Stemmer](https://milvus.io/docs/stemmer-filter.md): The stemmer filter reduces words to their base or root form (known as stemming), making it easier to match words with similar meanings across different… - [Remove Punct](https://milvus.io/docs/removepunct-filter.md): The removepunct filter removes tokens that contain punctuation or whitespace. - [Regex](https://milvus.io/docs/regex-filter.md): The regex analyzer filter keeps tokens that match a regular expression and discards the rest. - [Synonym](https://milvus.io/docs/synonym-filter.md): Use the synonym filter to rewrite tokens with a synonym dictionary during text analysis. #### Best Practices - [Data Model Design for Search](https://milvus.io/docs/schema-hands-on.md): Information Retrieval systems, also known as search engines, are essential for various AI applications such as Retrieval-augmented generation (RAG), visual… - [Data Model Design with an Array of Structs](https://milvus.io/docs/best-practices-for-array-of-structs.md): Modern AI applications, especially in the Internet of Things (IoT) and autonomous driving, typically reason over rich, structured events: a sensor reading with… ### Insert & Delete Use `upsert()` to modify existing entities; `update()` does not exist. - [Insert Entities](https://milvus.io/docs/insert-update-delete.md): Entities in a collection are data records that share the same set of fields. - [Upsert Entities](https://milvus.io/docs/upsert-entities.md): The upsert operation provides a convenient way to insert or update entities in a collection. - [Delete Entities](https://milvus.io/docs/delete-entities.md): You can delete the entities that are no longer needed by filtering conditions or their primary keys. ### Indexes A vector field must be indexed before the collection can be loaded and searched. AUTOINDEX is the right default for most workloads. - [Index Explained](https://milvus.io/docs/index-explained.md): An index is an additional structure built on top of data. - [FAISS](https://milvus.io/docs/faiss.md): Use FAISS index passthrough to supply Faiss index-factory strings and factory-specific search parameters in Milvus 3.0. #### Floating Vector Indexes - [FLAT](https://milvus.io/docs/flat.md): The FLAT index is one of the simplest and most straightforward methods for indexing and searching floating-point vectors. - [IVF_FLAT](https://milvus.io/docs/ivf-flat.md): The IVF_FLAT index is an indexing algorithm that can improve search performance for floating-point vectors. - [IVF_SQ8](https://milvus.io/docs/ivf-sq8.md): The IVF_SQ8 index is a quantization-based indexing algorithm designed to tackle large-scale similarity search challenges. - [IVF_PQ](https://milvus.io/docs/ivf-pq.md): The IVF_PQ index is a quantization-based indexing algorithm for approximate nearest neighbor search in high-dimensional spaces. - [IVF_RABITQ](https://milvus.io/docs/ivf-rabitq.md): The IVF_RABITQ index is a binary quantization-based indexing algorithm that quantizes FP32 vectors into binary representations. - [HNSW](https://milvus.io/docs/hnsw.md): The HNSW index is a graph-based indexing algorithm that can improve performance when searching for high-dimensional floating vectors. - [HNSW_SQ](https://milvus.io/docs/hnsw-sq.md): HNSW_SQ combines Hierarchical Navigable Small World (HNSW) graphs with Scalar Quantization (SQ), creating an advanced vector indexing method that offers a… - [HNSW_PQ](https://milvus.io/docs/hnsw-pq.md): HNSW_PQ leverages Hierarchical Navigable Small World (HNSW) graphs with Product Quantization (PQ), creating an advanced vector indexing method that offers a… - [HNSW_PRQ](https://milvus.io/docs/hnsw-prq.md): HNSW_PRQ leverages Hierarchical Navigable Small World (HNSW) graphs with Product Residual Quantization (PRQ), offering an advanced vector indexing method that… - [DISKANN](https://milvus.io/docs/diskann.md): In large-scale scenarios, where datasets can include billions or even trillions of vectors, standard in-memory indexing methods (e.g., HNSW, IVF_FLAT) often… - [SCANN](https://milvus.io/docs/scann.md): Powered by the ScaNN library from Google, the SCANN index in Milvus is designed to address scaling vector similarity search challenges, striking a balance… - [AISAQ](https://milvus.io/docs/aisaq.md): AISAQ is a disk-based vector index that extends DISKANN to handle billion-scale datasets without exceeding RAM limits. #### Binary Vector Indexes - [BIN_IVF_FLAT](https://milvus.io/docs/bin-ivf-flat.md): The BIN_IVF_FLAT index is a variant of the IVF_FLAT index exclusively for binary embeddings. - [BIN_FLAT](https://milvus.io/docs/bin-flat.md): The BIN_FLAT index is a variant of the FLAT index tailored exclusively for binary embeddings. - [MINHASH_LSH](https://milvus.io/docs/minhash-lsh.md): Use MinHash LSH indexes to speed up near-duplicate detection and Jaccard similarity search on large text datasets. #### Sparse Vector Indexes - [SPARSE_INVERTED_INDEX](https://milvus.io/docs/sparse-inverted-index.md): The SPARSE_INVERTED_INDEX index is an index type used by Milvus to efficiently store and search sparse vectors. #### Scalar Indexes - [BITMAP](https://milvus.io/docs/bitmap.md): Bitmap indexing is an efficient indexing technique designed to improve query performance on low-cardinality scalar fields. - [INVERTED](https://milvus.io/docs/inverted.md): When you need to perform frequent filter queries on your data, INVERTED indexes can dramatically improve query performance. - [NGRAM](https://milvus.io/docs/ngram.md): The NGRAM index in Milvus accelerates LIKE queries and eligible regex filters on VARCHAR fields or specific JSON paths within JSON fields. - [RTREE](https://milvus.io/docs/rtree.md): The RTREE index is a tree-based data structure that accelerates queries on GEOMETRY fields in Milvus. - [STL_SORT](https://milvus.io/docs/stl-sort.md): The STL_SORT index is an index type specifically designed to enhance query performance on numeric fields (INT8, INT16, etc.), VARCHAR fields, or TIMESTAMPTZ… #### GPU-enabled Indexes - [GPU Index Overview](https://milvus.io/docs/gpu-index-overview.md): Building an index with GPU support in Milvus can significantly improve search performance in high-throughput and high-recall scenarios. - [GPU_CAGRA](https://milvus.io/docs/gpu-cagra.md): The GPU_CAGRA index is a graph-based index optimized for GPUs. - [GPU_IVF_FLAT](https://milvus.io/docs/gpu-ivf-flat.md): The GPU_IVF_FLAT index is a GPU-accelerated version of the IVF_FLAT index, designed exclusively for GPU environments. - [GPU_IVF_PQ](https://milvus.io/docs/gpu-ivf-pq.md): The GPU_IVF_PQ index builds on the IVF_PQ concept by combining inverted file clustering with Product Quantization (PQ), which breaks down high-dimensional… - [GPU_BRUTE_FORCE](https://milvus.io/docs/gpu-brute-force.md): Dedicated to GPU environments, the GPU_BRUTE_FORCE index is engineered for scenarios where uncompromising accuracy is essential. ### Search Load the collection before searching. Hybrid search has per-request constraints — see the rules above. - [Basic ANN Search](https://milvus.io/docs/single-vector-search.md): Run basic ANN searches in Milvus with query vectors, output fields, filters, ranges, and iterators. - [Filtered Search](https://milvus.io/docs/filtered-search.md): An ANN search finds vector embeddings most similar to specified vector embeddings. - [Range Search](https://milvus.io/docs/range-search.md): A range search improves search result relevancy by restricting the distance or score of the returned entities within a specific range. - [Grouping Search](https://milvus.io/docs/grouping-search.md): Use grouping search to aggregate ANN search results by a field value and reduce duplicate entities. - [Search Aggregation](https://milvus.io/docs/search-aggregation.md): Group vector search results into buckets, compute per-bucket metrics, order buckets, and return representative hits. - [Primary-Key Search](https://milvus.io/docs/primary-key-search.md): When conducting similarity searches, you are always asked to provide one or more query vectors, even if the query vectors are already present in the target… - [Hybrid Search](https://milvus.io/docs/multi-vector-search.md): In many applications, an object can be searched by a rich set of information such as title and description, or with multiple modalities such as text, images,… - [Query](https://milvus.io/docs/get-and-scalar-query.md): Use Query, Get, and QueryIterator to retrieve entities, filter metadata, sort query results, and aggregate scalar values in Milvus. - [Full Text Search](https://milvus.io/docs/full-text-search.md): Full text search is a feature that retrieves documents containing specific terms or phrases in text datasets, then ranking the results based on relevance. - [Text Match](https://milvus.io/docs/keyword-match.md): Text match in Milvus enables precise document retrieval based on specific terms. - [Text Highlighter](https://milvus.io/docs/text-highlighter.md): The Highlighter in Milvus annotates matched terms in text fields by wrapping them with customizable tags. - [Phrase Match](https://milvus.io/docs/phrase-match.md): Phrase match lets you search for documents containing your query terms as an exact phrase. - [Elasticsearch Queries to Milvus](https://milvus.io/docs/elasticsearch-queries-to-milvus.md): Elasticsearch, built on Apache Lucene, is a leading open-source search engine. - [Search Iterators](https://milvus.io/docs/with-iterators.md): The ANN Search has a maximum limit on the number of entities that can be recalled in a single query, and simply using basic ANN Search may not meet the demands… - [Use Partition Key](https://milvus.io/docs/use-partition-key.md): The Partition Key is a search optimization solution based on partitions. #### Filtering - [Filtering Explained](https://milvus.io/docs/boolean.md): Milvus provides powerful filtering capabilities that enable precise querying of your data. - [Basic Operators](https://milvus.io/docs/basic-operators.md): Milvus provides a rich set of basic operators to help you filter and query data efficiently. - [Pattern Matching](https://milvus.io/docs/pattern-matching.md): Milvus supports string pattern matching with LIKE wildcard patterns and RE2 regular expressions. - [Filtering Templating](https://milvus.io/docs/filtering-templating.md): In Milvus, complex filter expressions with numerous elements, especially those involving non-ASCII characters like CJK characters, can significantly affect… - [JSON Operators](https://milvus.io/docs/json-operators.md): Milvus supports advanced operators for querying and filtering JSON fields, making them perfect for managing complex, structured data. - [Array Operators](https://milvus.io/docs/array-operators.md): Milvus provides ARRAY operators for filtering ARRAY fields and partially updating ARRAY field values. - [StructArray Operators](https://milvus.io/docs/struct-array-operators.md): StructArray operators filter entities by evaluating predicates on scalar subfields inside a StructArray field. - [Random Sampling](https://milvus.io/docs/random-sampling.md): When working with large-scale datasets, you often don't need to process all your data to gain insights or test filtering logic. - [Geometry Operators](https://milvus.io/docs/geometry-operators.md): Milvus supports a set of operators for spatial filtering on GEOMETRY fields, which are essential for managing and analyzing geometric data. - [Membership Match](https://milvus.io/docs/membership-match.md): Filter entities against large member sets using compact Bloom filters or exact Roaring bitmaps, with syntax, SDK request patterns, delete semantics, and… #### Search with StructArray - [Basic Vector Search with StructArray](https://milvus.io/docs/basic-vector-search-with-structarray.md): Use this page to run vector search on vector subfields inside a StructArray field. - [Filtered Search with StructArray](https://milvus.io/docs/filtered-search-with-structarray.md): Use this page to add scalar filtering to vector search on StructArray fields. - [Range Search with StructArray](https://milvus.io/docs/range-search-with-structarray.md): Use this page to run range search on StructArray vector subfields. - [Grouping Search with StructArray](https://milvus.io/docs/grouping-search-with-structarray.md): Use this page to group StructArray element-level search results by the parent entity. - [Hybrid Search with StructArray](https://milvus.io/docs/hybrid-search-with-structarray.md): Use this page to combine StructArray vector search with other vector searches in one hybrid search request. - [Search with EmbeddingLists: ColBERT and ColPali](https://milvus.io/docs/search-with-embedding-lists.md): This tutorial shows how to build ColBERT-style and ColPali-style retrieval systems with EmbeddingList search on StructArray vector subfields in Milvus. - [Choose an EmbeddingList Search Strategy](https://milvus.io/docs/choose-an-embeddinglist-search-strategy.md): EmbeddingList search strategies decide how Milvus builds an approximate candidate index for EmbeddingList search. ### Function & Model Inference - [BM25 Function](https://milvus.io/docs/bm25-function.md): The BM25 function enables full text search by transforming raw text into sparse vectors and scoring documents based on lexical relevance. - [MinHash Function](https://milvus.io/docs/minhash-function.md): Use MinHash to convert text into binary vectors for Jaccard-based similarity search and near-duplicate detection. #### Model-based Embedding Functions - [Embedding Function Overview](https://milvus.io/docs/embedding-function-overview.md): The Function module in Milvus allows you to transform raw text data into vector embeddings by automatically calling external embedding service providers (like… - [OpenAI](https://milvus.io/docs/openai.md): Use an OpenAI embedding model with Milvus by choosing a model and configuring Milvus with your OpenAI API key. - [Azure OpenAI](https://milvus.io/docs/azure-openai.md): This topic describes how to configure and use Azure OpenAI embedding functions in Milvus. - [DashScope](https://milvus.io/docs/dashscope.md): This topic describes how to configure and use DashScope embedding functions in Milvus. - [Bedrock](https://milvus.io/docs/bedrock.md): This topic describes how to configure and use Amazon Bedrock embedding functions in Milvus. - [Vertex AI](https://milvus.io/docs/vertex-ai.md): Google Cloud Vertex AI is a high-performance service specifically designed for text embedding models. - [Voyage AI](https://milvus.io/docs/voyage-ai.md): This topic describes how to configure and use Voyage AI embedding functions in Milvus. - [Cohere](https://milvus.io/docs/cohere.md): This topic describes how to configure and use Cohere embedding functions in Milvus. - [SiliconFLow](https://milvus.io/docs/siliconflow.md): This topic describes how to configure and use SiliconFLow embedding functions in Milvus. - [Hugging Face TEI](https://milvus.io/docs/hugging-face-tei.md): Hugging Face Text Embeddings Inference (TEI) is a high-performance inference server specifically designed for text embedding models. - [Hugging Face](https://milvus.io/docs/hugging-face.md): This topic describes how to use hosted Hugging Face Inference Providers for text embedding in Milvus. - [Google Gemini](https://milvus.io/docs/google-gemini.md): Use a Google Gemini embedding model with Milvus by choosing a model and configuring Milvus with your Gemini API key. - [Yandex Cloud](https://milvus.io/docs/yandex-cloud.md): This topic describes how to configure and use Yandex Cloud embedding functions in Milvus. #### Rerank Functions - [Weighted Ranker](https://milvus.io/docs/weighted-ranker.md): Weighted Ranker intelligently combines and prioritizes results from multiple search paths by assigning different importance weights to each. - [RRF Ranker](https://milvus.io/docs/rrf-ranker.md): Reciprocal Rank Fusion (RRF) Ranker is a reranking strategy for Milvus hybrid search that balances results from multiple vector search paths based on their… - [Boost Ranker](https://milvus.io/docs/boost-ranker.md): Instead of relying solely on semantic similarity calculated based on vector distances, Boost Rankers allow you to influence search results in a meaningful way. ##### Decay Ranker - [Decay Ranker Overview](https://milvus.io/docs/decay-ranker-overview.md): In traditional vector search, results are ranked purely by vector similarity—how closely vectors match in mathematical space. - [Gaussian Decay](https://milvus.io/docs/gaussian-decay.md): Gaussian decay, also known as normal decay, creates the most natural-feeling adjustment to your search results. - [Exponential Decay](https://milvus.io/docs/exponential-decay.md): Exponential decay creates a steep initial drop followed by a long tail in your search results. - [Linear Decay](https://milvus.io/docs/linear-decay.md): Linear decay creates a straight-line decline that terminates at an absolute zero point in your search results. - [Tutorial: Implement Time-based Ranking in Milvus](https://milvus.io/docs/tutorial-implement-a-time-based-ranking-in-milvus.md): In many search applications, the freshness of content is just as important as its relevance. ##### Model Ranker - [Model Ranker Overview](https://milvus.io/docs/model-ranker-overview.md): Traditional vector search ranks results purely by mathematical similarity—how closely vectors match in high-dimensional space. - [vLLM Ranker](https://milvus.io/docs/vllm-ranker.md): The vLLM Ranker leverages the vLLM inference framework to enhance search relevance through semantic reranking. - [TEI Ranker](https://milvus.io/docs/tei-ranker.md): The TEI Ranker leverages the Text Embedding Inference (TEI) service from Hugging Face to enhance search relevance through semantic reranking. - [Cohere Ranker](https://milvus.io/docs/cohere-ranker.md): The Cohere Ranker leverages Cohere's powerful rerank models to enhance search relevance through semantic reranking. - [Voyage AI Ranker](https://milvus.io/docs/voyage-ai-ranker.md): The Voyage AI Ranker leverages Voyage AI's specialized rerankers to enhance search relevance through semantic reranking. - [SiliconFlow Ranker](https://milvus.io/docs/siliconflow-ranker.md): The SiliconFlow Ranker leverages SiliconFlow's comprehensive reranking models to enhance search relevance through semantic reranking. - [DashScope Ranker](https://milvus.io/docs/dashscope-ranker.md): This topic describes how to configure and use DashScope reranking models, such as Qwen rerank models, in Milvus. - [Hugging Face Ranker](https://milvus.io/docs/hugging-face-ranker.md): This topic describes how to rerank Milvus search results with hosted Hugging Face sentence-similarity models. ### Storage Optimization - [Use mmap](https://milvus.io/docs/mmap.md): Memory mapping (Mmap) enables direct memory access to large files on disk, allowing Milvus to store indexes and data in both memory and hard drives. - [Clustering Compaction](https://milvus.io/docs/clustering-compaction.md): Clustering compaction is designed to improve search performance and reduce costs in large collections. - [Force Merge Compaction](https://milvus.io/docs/force-merge.md): Use force merge compaction to consolidate small segments and improve query performance and storage efficiency. #### Tiered Storage - [Tiered Storage Overview](https://milvus.io/docs/tiered-storage-overview.md): In Milvus, the traditional full-load mode requires each QueryNode to load all data fields and indexes of a segment at initialization, even data that may never… - [Warm Up](https://milvus.io/docs/warm-up.md): Warm Up complements Tiered Storage by preloading selected fields or indexes into the cache before a segment becomes queryable. - [Eviction](https://milvus.io/docs/eviction.md): Eviction manages the cache resources of each QueryNode in Milvus. ### Snapshots - [Overview](https://milvus.io/docs/snapshots.md): Use snapshots to capture point-in-time collection states for rollback, versioning, and testing. - [Manage Snapshots](https://milvus.io/docs/manage-snapshots.md): Learn how to create, list, describe, pin, restore, and drop snapshots and monitor restoration jobs. - [Use Cases](https://milvus.io/docs/snapshot-use-cases.md): In this guide, you will find common use cases for snapshots. ## Data Import - [Prepare Source Data](https://milvus.io/docs/prepare-source-data.md): This page discusses something you should consider before you start bulk-inserting data into your collection. - [Import Data](https://milvus.io/docs/import-data.md): This page demonstrates the procedure to import the prepared data. ## AI Tools Written specifically for AI coding agents. `agents_overview.md` is a drop-in AGENTS.md for Milvus; read it before generating Milvus code. - [Milvus for AI Agents](https://milvus.io/docs/milvus_for_agents.md): Learn how AI agents can use Milvus as a vector database for RAG, semantic search, and long-term memory. ### AI Prompts - [AGENTS.md for Milvus](https://milvus.io/docs/agents_overview.md): Rules and patterns for AI coding agents that generate, review, or debug Milvus vector database code using PyMilvus. - [Python SDK](https://milvus.io/docs/python_sdk.md): Rules for AI coding assistants to write correct Milvus Python code using MilvusClient. - [Schema Design](https://milvus.io/docs/schema_design.md): Rules for AI coding assistants to design correct Milvus collection schemas. - [Search Patterns](https://milvus.io/docs/search_patterns.md): Rules for AI coding assistants to implement search, hybrid search, and full-text search in Milvus. - [Index Selection](https://milvus.io/docs/index_selection.md): Rules for AI coding assistants to choose and configure Milvus indexes. - [RAG Pipeline](https://milvus.io/docs/rag_pipeline.md): Rules for AI coding assistants to build RAG pipelines with Milvus. ## Administration Guide Deploying, configuring, scaling, monitoring, and securing Milvus in production. Downgrading across versions is not supported. - [Milvus WebUI](https://milvus.io/docs/milvus-webui.md): Milvus Web UI is a graphical management tool for Milvus. It enhances system observability with a simple and intuitive interface. ### Deployment #### On Premises - [With Docker](https://milvus.io/docs/configure-docker.md): Configure Milvus with Docker Compose. - [With Helm](https://milvus.io/docs/configure-helm.md): Configure Milvus with Helm Charts. - [With Milvus Operator](https://milvus.io/docs/configure_operator.md): Learn how to configure Milvus with Milvus Operator. - [Allocate Resources](https://milvus.io/docs/allocate.md): Learn how to allocate resources to Milvus on Kubernetes. #### On Clouds - [Deploy on AWS](https://milvus.io/docs/eks.md): Learn how to deploy a Milvus cluster on EKS - [Deploy on GCP](https://milvus.io/docs/gcp.md): Learn how to deploy a Milvus cluster on GKE. - [Deploy on Azure](https://milvus.io/docs/azure.md): Learn how to deploy a Milvus cluster on Azure. - [Deploy on OpenShift](https://milvus.io/docs/openshift.md): Learn how to deploy a Milvus cluster on OpenShift. ##### Layer-7 Load Balacing - [AWS](https://milvus.io/docs/aws_layer7.md): Learn how to deploy a Milvus cluster behind a Layer-7 load balancer on AWS. - [GCP](https://milvus.io/docs/gcp_layer7.md): Learn how to deploy a Milvus cluster behind a Layer-7 load balancer on GCP. - [Azure](https://milvus.io/docs/ingress.md): Learn how to configure ingress nginx with Milvus. ##### Cloud Storage - [AWS](https://milvus.io/docs/s3.md): Learn how to configure s3 with IAM Role. - [GCP](https://milvus.io/docs/gcs.md): Learn how to configure gcs with Workload Identity. - [Azure](https://milvus.io/docs/abs.md): Learn how to configure Blob Storage with Workload Identity. ### Configuration - [System Configurations](https://milvus.io/docs/system_configuration.md): Learn about the system configuration of Milvus. - [On the Fly](https://milvus.io/docs/dynamic_config.md): Learn about the dynamic configuration of Milvus. - [Limit Collection Counts](https://milvus.io/docs/limit_collection_counts.md): Configure the maximum number of collections in an instance. - [Configure Chunk Cache](https://milvus.io/docs/chunk_cache.md): Tune the chunk cache that pre-loads data into memory ahead of search. - [Coordinator HA](https://milvus.io/docs/coordinator_ha.md): Learn about the motivation and procedure for Milvus coordinators to work in active standby. - [QueryNode Using Local Disk](https://milvus.io/docs/configure-querynode-localdisk.md): Learn how to configure Milvus QueryNode to use local disk. ### Data Infrastructure - [Overview](https://milvus.io/docs/data-infra-integration-overview.md): Overview of the metadata storage, object storage, and message queues that Milvus uses. - [Object Storage](https://milvus.io/docs/object-storage.md) #### Metadata - [etcd](https://milvus.io/docs/etcd.md) #### Message Queue - [Overview](https://milvus.io/docs/mqtype-overview.md): Overview of the message queue (mqType) options Milvus supports, and which one to use for standalone vs. distributed deployments. - [Woodpecker](https://milvus.io/docs/woodpecker.md): Learn how Woodpecker works as the default message queue (WAL) in Milvus, and how to run it in embedded or service mode. - [Pulsar](https://milvus.io/docs/mq_pulsar.md) - [Kafka](https://milvus.io/docs/mq_kafka.md) - [RocksMQ](https://milvus.io/docs/mq_rocksmq.md) ##### Pulsar - [Upgrade to Pulsar v3](https://milvus.io/docs/upgrade-pulsar-v3.md): Learn how to upgrade Pulsar from V2 to V3 in Milvus so that you can use the latest version of Milvus v2.5.x. - [Continue Using Pulsar v2](https://milvus.io/docs/use-pulsar-v2.md): Milvus recommands you to upgrade Pulsar to v3 for Milvus v2.5.x. ##### Switch Message Queue - [Overview](https://milvus.io/docs/switch-mq-type.md): Switch an existing Milvus deployment between Woodpecker and another message queue without downtime. - [RocksMQ ↔ Woodpecker](https://milvus.io/docs/switch-rocksmq-woodpecker.md): Switch the message queue of a Milvus Standalone (Docker Compose) deployment between RocksMQ and Woodpecker. - [Pulsar ↔ Woodpecker](https://milvus.io/docs/switch-pulsar-woodpecker.md): Switch the message queue of a Milvus cluster between Pulsar and Woodpecker, with Helm or Milvus Operator. - [Kafka ↔ Woodpecker](https://milvus.io/docs/switch-kafka-woodpecker.md): Switch the message queue of a Milvus cluster between Kafka and Woodpecker, with Helm or Milvus Operator. ### Scaling - [Scale Cluster](https://milvus.io/docs/scaleout.md): Learn how to manually or automatically scale out and scale in a Milvus cluster. - [Scale Standalone](https://milvus.io/docs/scale-standalone.md): Milvus Standalone is a single-machine server deployment. - [Scale with HPA](https://milvus.io/docs/hpa.md): Learn how to configure Horizontal Pod Autoscaling (HPA) to dynamically scale a Milvus cluster. - [Scale Dependencies](https://milvus.io/docs/scale-dependencies.md): Scale the etcd, object storage, and message queue dependencies. ### Upgrade - [Upgrade Milvus Cluster](https://milvus.io/docs/upgrade_milvus_cluster-operator.md): Learn how to upgrade Milvus cluster with Milvus Operator. - [Upgrade Milvus Standalone](https://milvus.io/docs/upgrade_milvus_standalone-operator.md): Learn how to upgrade Milvus standalone with Milvus Operator. ### Monitoring, Alerts & Logs #### Monitoring - [Architecture](https://milvus.io/docs/monitor_overview.md): Learn how Prometheus and Grafana are used in Milvus for montoring and alerting services. - [Deploy Monitoring Services](https://milvus.io/docs/monitor.md): Learn how to deploy monitoring services for a Milvus cluster using Prometheus. - [Visualize Metrics](https://milvus.io/docs/visualize.md): Learn how to visualize Milvus metrics in Grafana. - [Milvus Metrics Dashboard](https://milvus.io/docs/metrics_dashboard.md): This topic introduces the monitoring metrics displayed in the Milvus Dashboard. #### Alerts - [Create an Alert](https://milvus.io/docs/alert.md): Learn how to create an alert for Milvus services in Grafana. #### Logs - [Configure Grafana Loki](https://milvus.io/docs/configure_grafana_loki.md): This topic describes how to collect logs using Loki and query logs for a Milvus cluster using Grafana. - [Configure Access Logs](https://milvus.io/docs/configure_access_logs.md): Enable and configure Milvus access logging. #### Trace - [Jaeger Tracing](https://milvus.io/docs/config_jaeger_tracing.md): This guide provides instructions on how to configure Jaeger to collect traces for Milvus. ### Resource Groups - [Manage Resource Groups](https://milvus.io/docs/resource_group.md): Learn how to manage resource groups. ### Security - [Enable Authentication](https://milvus.io/docs/authenticate.md): Learn how to manage user authentication in Milvus. - [Encryption in Transit](https://milvus.io/docs/tls.md): Learn how to enable TLS proxy in Milvus. - [Connect to Kafka with SASL/SSL](https://milvus.io/docs/connect_kafka_ssl.md): This guide lists several ways to connect Milvus to Kafka, from the simplest one without SASL/SSL to the fully secured one with SASL/SSL. #### Enable RBAC - [RBAC Overview](https://milvus.io/docs/rbac.md): RBAC (Role-Based Access Control) is an access control method based on roles. - [Users, Privileges, and Roles](https://milvus.io/docs/users_and_roles.md): Milvus achieves fine-grained access control through RBAC. - [Privilege Group](https://milvus.io/docs/privilege_group.md): To streamline the process of granting privileges, it is recommended that you combine multiple privileges into a privilege group. - [Grant Privileges](https://milvus.io/docs/grant_privileges.md): Once a role is created, you can grant privileges to the role. This guide introduces how to grant privileges or privilege groups to a role. - [Grant Roles](https://milvus.io/docs/grant_roles.md): After creating a role and granting privileges to the role, you can grant the role to users so that the users can access resources and perform actions that are… - [Drop Users & Roles](https://milvus.io/docs/drop_users_roles.md): To ensure data security, it is recommend that you drop users and roles that are no longer in use. This guide introduces how to drop users and roles. ## Tools - [SDK Code Helper (MCP)](https://milvus.io/docs/milvus-sdk-helper-mcp.md): ⚡️ Configure once, boost efficiency forever! ### Milvus Backup - [Overview](https://milvus.io/docs/milvus_backup_overview.md): Milvus-Backup is a tool that allows users to backup and restore Milvus data. - [Commands](https://milvus.io/docs/milvus_backup_cli.md): Learn how to use Milvus Backup through CLI - [RESTful API](https://milvus.io/docs/milvus_backup_api.md): Learn how to use Milvus Backup through API #### Common Cases - [Backup and Restore in One Instance](https://milvus.io/docs/single-instance-backup-and-restore.md): This topic details the process of backing up a collection and restoring it from the backup within the same Milvus instance - [Migrate Between Instances in One Bucket](https://milvus.io/docs/shared-bucket-backup-and-restore.md): This topic details the process of backing up a collection from one Milvus instance and restoring it to another while using a shared bucket for object storage - [Migrate Between Instances Across Buckets](https://milvus.io/docs/cross-bucket-backup-and-restore.md): This topic details the process of backing up a collection from one Milvus instance and restoring it to another - [Migrate Between Instances Across S3 Environments](https://milvus.io/docs/multi-storage-backup-and-restore.md): This topic details the process of backing up a collection from one Milvus instance and restoring it to another ### Birdwatcher - [Overview](https://milvus.io/docs/birdwatcher_overview.md): Birdwatcher is a debug tool for Milvus 2.x. It connects to etcd and inspects the status of the Milvus system. - [Install Guides](https://milvus.io/docs/birdwatcher_install_guides.md): Learn how to install Birdwatch to debug Milvus. - [Usage Guides](https://milvus.io/docs/birdwatcher_usage_guides.md): Learn how to use Birdwatch to debug Milvus. ### Milvus CDC - [Overview](https://milvus.io/docs/milvus_cdc_overview.md): Milvus CDC replicates data changes from one Milvus cluster to another for primary-standby disaster recovery. - [Set Up CDC Replication](https://milvus.io/docs/set_up_cdc_replication.md): Learn how to deploy two Milvus clusters and configure CDC replication between them. - [Switchover](https://milvus.io/docs/cdc_switchover.md): Learn how to perform a planned switchover between primary and standby Milvus clusters with Milvus CDC. - [Failover](https://milvus.io/docs/cdc_failover.md): Learn how to perform a failover when the primary Milvus cluster becomes unavailable. - [Bulk Import in CDC Replication](https://milvus.io/docs/bulk_import_in_cdc_replication.md): Learn how to run a bulk import against Milvus clusters that use CDC replication. ### Milvus CLI - [Overview](https://milvus.io/docs/cli_overview.md): Milvus Command-Line Interface (CLI) is a command-line tool that supports database connection, data operations, and import and export of data. - [Installation](https://milvus.io/docs/install_cli.md): Learn how to install Milvus_CLI. - [Commands](https://milvus.io/docs/cli_commands.md): Interact with Milvus using commands. ### Milvus Connectors - [Spark](https://milvus.io/docs/integrate_with_spark.md): Apache Spark and Databricks integrates with Milvus and Zilliz Cloud to combine big data processing with vector search. ### Models #### Embeddings - [Overview](https://milvus.io/docs/embeddings.md): Learn how to generate embeddings for your data. - [OpenAI](https://milvus.io/docs/embed-with-openai.md): Milvus integrates with OpenAI's models via the OpenAIEmbeddingFunction class. - [Sentence Transformers](https://milvus.io/docs/embed-with-sentence-transform.md): This article demonstrates how to use Sentence Transformers in Milvus to encode documents and queries into dense vectors. - [BGE M3](https://milvus.io/docs/embed-with-bgm-m3.md): BGE-M3 is named for its capabilities in Multi-Linguality, Multi-Functionality, and Multi-Granularity. - [SPLADE](https://milvus.io/docs/embed-with-splade.md): This article describes how to use the SpladeEmbeddingFunction to encode documents and queries using the SPLADE model. - [Voyage](https://milvus.io/docs/embed-with-voyage.md): This article describes how to use the VoyageEmbeddingFunction to encode documents and queries using the Voyage model. - [Jina AI](https://milvus.io/docs/embed-with-jina.md): This article describes how to use the JinaEmbeddingFunction to encode documents and queries using the Jina AI embedding model. - [Cohere](https://milvus.io/docs/embed-with-cohere.md): This article describes how to use the CohereEmbeddingFunction to encode documents and queries using the Cohere embedding model. - [Instructor](https://milvus.io/docs/embed-with-instructor.md): This article describes how to use the InstructorEmbeddingFunction to encode documents and queries using the Instructor embedding model. - [Mistral AI](https://milvus.io/docs/embed-with-mistral-ai.md): This article describes how to use the MistralAIEmbeddingFunction to encode documents and queries using the Mistral AI embedding model. - [Nomic](https://milvus.io/docs/embed-with-nomic.md): This article describes how to use the NomicEmbeddingFunction to encode documents and queries using the Nomic embedding model. - [mGTE](https://milvus.io/docs/embed-with-mgte.md): This article describes how to use the MGTEEmbeddingFunction to encode documents and queries using the mGTE embedding model. - [Model2Vec](https://milvus.io/docs/embed-with-model2vec.md): Milvus integrates with Model2Vec's models via the Model2VecEmbeddingFunction class. - [Gemini](https://milvus.io/docs/embed-with-gemini.md): Milvus integrates with OpenAI's models via the GeminiEmbeddingFunction class. #### Rerankers - [Overview](https://milvus.io/docs/rerankers-overview.md): PyMilvus model library integrates rerank functions to optimize the order of results returned from initial searches. - [BGE](https://milvus.io/docs/rerankers-bge.md): Milvus supports BGE reranker models through the `BGERerankFunction` class. - [Cohere](https://milvus.io/docs/rerankers-cohere.md): Milvus supports Cohere reranker models through the `CohereRerankFunction` class. - [Cross Encoder](https://milvus.io/docs/rerankers-cross-encoder.md): Milvus supports Cross Encoder reranker models through the `CrossEncoderRerankFunction` class. - [Voyage](https://milvus.io/docs/rerankers-voyage.md): Milvus supports Voyage reranker model through the `VoyageRerankFunction` class. - [Jina AI](https://milvus.io/docs/rerankers-jina.md): Milvus supports Jina reranker model through the `JinaRerankFunction` class. ## Integrations - [Overview](https://milvus.io/docs/integrations_overview.md): This page provides a list of tutorials for you to interact with Milvus and third-party tools. ### Orchestration - [DSPy](https://milvus.io/docs/integrate_with_dspy.md): This guide demonstrates how to use MilvusRM, one of DSPy's retriever modules, to optimize RAG programs. - [FastGPT](https://milvus.io/docs/integrate_with_fastgpt.md): This tutorial will guide you on how to swiftly deploy your own exclusive FastGPT application using Milvus. - [Kotaemon](https://milvus.io/docs/kotaemon_with_milvus.md): This tutorial will guide you on how to customize your kotaemon application using Milvus. - [Dify](https://milvus.io/docs/dify_with_milvus.md): In this tutorial, we will show you how to deploy Dify with Milvus, to enable efficient retrieval and RAG engine. - [Langflow](https://milvus.io/docs/rag_with_langflow.md): This guide demonstrates how to use Langflow to build a Retrieval-Augmented Generation (RAG) pipeline with Milvus. - [DocsGPT](https://milvus.io/docs/use_milvus_in_docsgpt.md): In this tutorial, we will show you how to use Milvus as the backend vector database for DocsGPT. - [PrivateGPT](https://milvus.io/docs/use_milvus_in_private_gpt.md): In this tutorial, we will show you how to use Milvus as the backend vector database for PrivateGPT. - [Dynamiq](https://milvus.io/docs/milvus_rag_with_dynamiq.md): In this tutorial, we’ll explore how to seamlessly use Dynamiq with Milvus, the high-performance vector database purpose-built for RAG workflows. - [Llama Stack](https://milvus.io/docs/llama_stack_with_milvus.md): This tutorial introduces how to build a Llama Stack Server configured with Milvus, enabling you to import your private data to serve as your knowledge base. - [n8n](https://milvus.io/docs/milvus_and_n8n.md): n8n is a powerful open-source workflow automation platform that allows you to connect various applications, services, and APIs together to create automated… - [AnythingLLM](https://milvus.io/docs/use_milvus_in_anythingllm.md): This guide will walk you through configuring Milvus as the vector database in AnythingLLM, enabling you to embed, store, and search your documents for… #### LangChain - [Basic Usage](https://milvus.io/docs/basic_usage_langchain.md): This notebook shows how to use functionality related to the Milvus vector database. - [Text Embedding](https://milvus.io/docs/langchain_milvus_dido.md): This guide demonstrates how to use Milvus 2.6's Text Embedding Function (also known as Data In Data Out) with LangChain. - [RAG](https://milvus.io/docs/integrate_with_langchain.md): This guide demonstrates how to build a Retrieval-Augmented Generation (RAG) system using LangChain and Milvus. - [Hybrid Search](https://milvus.io/docs/milvus_hybrid_search_retriever.md): This notebook shows how to use functionality related to the Milvus vector database. - [Full-text Search](https://milvus.io/docs/full_text_search_with_langchain.md): This tutorial will show how to use LangChain and Milvus to implement full-text search in your application. - [Asynchronous Search](https://milvus.io/docs/langchain_milvus_async.md): This tutorial explores how to leverage asynchronous functions in langchain-milvus to build high-performance applications. - [Agent RAG](https://milvus.io/docs/agentic_rag_with_milvus_and_langgraph.md): This guide demonstrates how to build an advanced Retrieval-Augmented Generation (RAG) system using LangGraph and Milvus. #### LlamaIndex - [RAG](https://milvus.io/docs/integrate_with_llamaindex.md): This guide demonstrates how to build a Retrieval-Augmented Generation (RAG) system using LlamaIndex and Milvus. - [Milvus Async API](https://milvus.io/docs/llamaindex_milvus_async.md): This tutorial demonstrates how to use LlamaIndex with Milvus to build asynchronous document processing pipeline for RAG. - [Full-text Search](https://milvus.io/docs/llamaindex_milvus_full_text_search.md): In this tutorial, you'll learn how to use LlamaIndex and Milvus to build a RAG system using full-text search and hybrid search. - [Hybrid Search](https://milvus.io/docs/llamaindex_milvus_hybrid_search.md): This notebook demonstrates how to use Milvus for hybrid search in LlamaIndex RAG pipelines. - [Metadata Filtering](https://milvus.io/docs/llamaindex_milvus_metadata_filter.md): This notebook illustrates the use of the Milvus vector store in LlamaIndex, focusing on metadata filtering capabilities. #### HayStack - [RAG](https://milvus.io/docs/integrate_with_haystack.md): This guide demonstrates how to build a Retrieval-Augmented Generation (RAG) system using Haystack and Milvus. - [Full-text Search](https://milvus.io/docs/full_text_search_with_milvus_and_haystack.md): This tutorial demonstrates how to implement full-text and hybrid search in your application using Haystack and Milvus. ### Agents - [MemGPT](https://milvus.io/docs/integrate_with_memgpt.md): MemGPT makes it easy to build and deploy stateful LLM agents. With Milvus integration, you can build agents with connections to external data sources (RAG). - [Camel](https://milvus.io/docs/integrate_with_camel.md): This guide demonstrates how to build a Retrieval-Augmented Generation (RAG) system using CAMEL and Milvus. - [Mem0](https://milvus.io/docs/quickstart_mem0_with_milvus.md): In this tutorial, we’ll cover essential Mem0 memory management operations—adding, retrieving, updating, searching, deleting, and tracking memory history—using… - [Agno](https://milvus.io/docs/integrate_with_agno.md): Milvus vector database enable efficient storage and retrieval of information as embeddings. - [OpenAI](https://milvus.io/docs/openai_agents_milvus.md): This notebook shows how to create an agent that can query Milvus using natural language through Function Calling. - [MCP](https://milvus.io/docs/milvus_and_mcp.md): This tutorial walks you through setting up an MCP server for Milvus, allowing AI applications to perform vector searches, manage collections, and retrieve data… - [MemPalace](https://milvus.io/docs/mempalace_with_milvus.md): In this tutorial, we will use the MemPalace CLI to mine a real subset of the public Milvus documentation and store it in Milvus. - [Google ADK](https://milvus.io/docs/google_adk_with_milvus.md): In this tutorial, we will use adk-milvus to connect ADK with Milvus in two common places a retrieval toolset over a knowledge base, and a cross-session memory… - [Basic Memory](https://milvus.io/docs/basic_memory_with_milvus.md): In this tutorial, we will build a small memory project for an application team. - [EverOS](https://milvus.io/docs/everos_with_milvus.md): In this tutorial, we will build a project assistant that remembers release decisions across separate conversations. ### Evaluation & Observability - [Ragas](https://milvus.io/docs/integrate_with_ragas.md): This guide demonstrates how to use Ragas to evaluate a Retrieval-Augmented Generation (RAG) pipeline built upon Milvus. - [LangFuse](https://milvus.io/docs/integrate_with_langfuse.md): This is a simple cookbook that demonstrates how to use the LlamaIndex Langfuse integration. It uses Milvus Lite to store the documents and Query. - [FiftyOne](https://milvus.io/docs/integrate_with_voxel51.md): This page discusses the integration with voxel51 - [Arize Pheonix](https://milvus.io/docs/evaluation_with_phoenix.md): This guide demonstrates how to use Arize Pheonix to evaluate a Retrieval-Augmented Generation (RAG) pipeline built upon Milvus. - [DeepEval](https://milvus.io/docs/evaluation_with_deepeval.md): This guide demonstrates how to use DeepEval to evaluate a Retrieval-Augmented Generation (RAG) pipeline built upon Milvus. - [AIMon](https://milvus.io/docs/AIMon_milvus_integration.md): In this tutorial, we'll help you build a retrieval-augmented generation (RAG) chatbot that answers questions. ### Embedding Models - [OpenAI](https://milvus.io/docs/integrate_with_openai.md): This page discusses vector database integration with OpenAI's embedding API. - [Cohere](https://milvus.io/docs/integrate_with_cohere.md): This page illustrates how to create a question-answering system based on the SQuAD dataset using Milvus as the vector database and Cohere as the embedding… - [HuggingFace](https://milvus.io/docs/integrate_with_hugging-face.md): This tutorial shows how to build a question answering system using Hugging Face as the data loader & embedding generator for data processing and Milvus as the… - [PyTorch](https://milvus.io/docs/integrate_with_pytorch.md): This page demostrates how to build image search with PyTorch and Milvus - [SentenceTransformers](https://milvus.io/docs/integrate_with_sentencetransformers.md): This page discusses movie search using Milvus - [VoyageAI](https://milvus.io/docs/integrate_with_voyageai.md): This page discusses vector database integration with VoyageAI's embedding API. - [Jina AI](https://milvus.io/docs/integrate_with_jina.md): This guide demonstrates how to use Jina embeddings and Milvus to conduct similarity search and retrieval tasks. - [Twelve Labs](https://milvus.io/docs/video_search_with_twelvelabs_and_milvus.md): Learn how to create a semantic video search application by integrating Twelve Labs' Embed API for generating multimodal embeddings with Milvus. - [BentoML](https://milvus.io/docs/integrate_with_bentoml.md): This guide demonstrates how to use an open-source embedding model and large-language model on BentoCloud with Milvus vector database to build a Retrieval… - [EmbedAnything](https://milvus.io/docs/build_RAG_with_milvus_and_embedAnything.md): In this tutorial, we’ll demonstrate how to build a Retrieval-Augmented Generation (RAG) pipeline using EmbedAnything together with Milvus. ### LLMs - [vLLM](https://milvus.io/docs/milvus_rag_with_vllm.md): This blog will show you how to build and run a RAG with Milvus, vLLM, and Llama 3.1. - [Mistral AI](https://milvus.io/docs/llama_agents_metadata.md): In this Notebook, we will explore different ideas: Store data into Milvus, use llama-index with Mistral models for data queries, create automated data search… - [Fireworks](https://milvus.io/docs/build_RAG_with_milvus_and_fireworks.md): In this tutorial, we will show you how to build a RAG (Retrieval-Augmented Generation) pipeline with Milvus and Fireworks AI. - [SiliconFlow](https://milvus.io/docs/build_RAG_with_milvus_and_siliconflow.md): In this tutorial, we will show you how to build a RAG(Retrieval-Augmented Generation) pipeline with Milvus and SiliconFlow. - [SambaNova](https://milvus.io/docs/use_milvus_with_sambanova.md): This tutorial leverages Milvus integration in SambaNova AI Starter Kits to build an Enterprise Knowledge Retrieval system, similar to RAG(Retrieval-Augmented… - [Gemini](https://milvus.io/docs/build_RAG_with_milvus_and_gemini.md): In this tutorial, we will show you how to build a RAG (Retrieval-Augmented Generation) pipeline with Milvus and Gemini. - [Ollama](https://milvus.io/docs/build_RAG_with_milvus_and_ollama.md): In this guide, we’ll show you how to leverage Ollama and Milvus to build a RAG (Retrieval-Augmented Generation) pipeline efficiently and securely. - [DeepSeek](https://milvus.io/docs/build_RAG_with_milvus_and_deepseek.md): In this tutorial, we’ll show you how to build a Retrieval-Augmented Generation (RAG) pipeline using Milvus and DeepSeek. ### Knowledge Engineering - [WhyHow](https://milvus.io/docs/integrate_with_whyhow.md): This guide demonstrates how to use whyhow.ai and Milvus Lite to conduct Rule-based Retrieval. - [Vanna](https://milvus.io/docs/integrate_with_vanna.md): This guide demonstrates how to use Vanna to generate and execute SQL queries based on your data stored in a database. - [Knowledge Table](https://milvus.io/docs/knowledge_table_with_milvus.md): By default, Knowledge Table uses the Milvus database to store and retrieve the extracted data. - [Cognee](https://milvus.io/docs/build_RAG_with_milvus_and_cognee.md): In this tutorial, we will show you how to build a RAG (Retrieval-Augmented Generation) pipeline with Milvus and Cognee. - [MindsDB](https://milvus.io/docs/integration_with_mindsdb.md): This tutorial demonstrates how to integrate Milvus with MindsDB, enabling you to leverage MindsDB's AI capabilities with Milvus's vector database functionality… ### Data Sources - [Airbyte](https://milvus.io/docs/integrate_with_airbyte.md): Airbyte is an open-source data movement infrastructure for building extract and load (EL) data pipelines. - [Kafka](https://milvus.io/docs/kafka-connect-milvus.md): Apache Kafka is integrated with Milvus and Zilliz Cloud to stream vector data. - [Apify](https://milvus.io/docs/apify_milvus_rag.md): This tutorial explains how to crawl websites using Apify's Website Content Crawler and save the data into Milvus/Zilliz vector database to be later used for… - [Unstructured](https://milvus.io/docs/rag_with_milvus_and_unstructured.md): In this tutorial, we will use Unstructured to ingest PDF documents and then use Milvus to build a RAG pipeline. - [PII Masker](https://milvus.io/docs/RAG_with_pii_and_milvus.md): In this tutorial, we will show you how to build a RAG(Retrieval-Augmented Generation) pipeline with Milvus and PII Masker. - [Firecrawl](https://milvus.io/docs/build_RAG_with_milvus_and_firecrawl.md): In this tutorial, we’ll show you how to build a Retrieval-Augmented Generation (RAG) pipeline using Milvus and Firecrawl. - [Crawl4AI](https://milvus.io/docs/build_RAG_with_milvus_and_crawl4ai.md): In this tutorial, we’ll show you how to build a Retrieval-Augmented Generation (RAG) pipeline using Milvus and Crawl4AI. - [AWS S3](https://milvus.io/docs/build_RAG_from_s3_with_milvus.md): This tutorial walks you through the process of building a Retrieval-Augmented Generation (RAG) pipeline using Milvus and Amazon S3. - [VectorETL](https://milvus.io/docs/ETL_using_vectorETL.md): In this tutorial, we'll explore how to efficiently load data into Milvus using VectorETL, a lightweight ETL framework designed for vector databases. - [Feast](https://milvus.io/docs/build_RAG_with_milvus_and_feast.md): In this tutorial, we will build a Retrieval-Augmented Generation (RAG) pipeline using Feast and Milvus. - [Mistral OCR](https://milvus.io/docs/mistral_ocr_with_milvus.md): This tutorial demonstrates how to build a document understanding system using Milvus and Mistral OCR. - [Docling](https://milvus.io/docs/build_RAG_with_milvus_and_docling.md): In this tutorial, we’ll show you how to build a Retrieval-Augmented Generation (RAG) pipeline using Milvus and Docling. - [NLWeb](https://milvus.io/docs/NLWeb_with_milvus.md): Learn how to integrate Microsoft NLWeb with Milvus to build powerful natural language interfaces for websites. - [LangExtract](https://milvus.io/docs/langextract_milvus_demo.md): This guide demonstrates how to use LangExtract with Milvus to build an intelligent document processing and retrieval system. - [Exa](https://milvus.io/docs/build_RAG_with_milvus_and_exa.md): This tutorial demonstrates how to build an agent that searches both the public web (via Exa) and a private knowledge base (via Milvus), then synthesizes a… ### Others - [Snowflake](https://milvus.io/docs/integrate_with_snowpark.md): This guide demonstrates how to start a Milvus demo on Snowpark container services. - [Arm](https://milvus.io/docs/build_rag_on_arm.md): In this tutorial, you learn how to build a Retrieval-Augmented Generation (RAG) application on Arm-based infrastructures. - [Jev](https://milvus.io/docs/search_with_jev.md): Vector search finds information related to a query. ## Tutorials - [Overview](https://milvus.io/docs/tutorials-overview.md): This page provides a list of tutorials for you to interact with Milvus. - [Build RAG with Milvus](https://milvus.io/docs/build-rag-with-milvus.md): build rag with milvus - [Advanced RAG](https://milvus.io/docs/how_to_enhance_your_rag.md): With the increasing popularity of Retrieval Augmented Generation RAG applications, there is a growing concern about improving their performance. - [Full-Text Search with Milvus](https://milvus.io/docs/full_text_search_with_milvus.md): Full-text search is a traditional method for retrieving documents by matching specific keywords or phrases in the text. - [Hybrid Search with Milvus](https://milvus.io/docs/hybrid_search_with_milvus.md): Hybrid Search with Milvus - [Image Search with Milvus](https://milvus.io/docs/image_similarity_search.md): image search with Milvus - [Multimodal RAG](https://milvus.io/docs/multimodal_rag_with_milvus.md): Multimodal RAG with Milvus - [Graph RAG with Milvus](https://milvus.io/docs/graph_rag_with_milvus.md): Graph RAG with Milvus - [Contextual Retrieval](https://milvus.io/docs/contextual_retrieval_with_milvus.md): Contextal Retrieval is an advanced retrieval method proposed by Anthropic to address the issue of semantic isolation of chunks, which arises in current… - [HDBSCAN Clustering](https://milvus.io/docs/hdbscan_clustering_with_milvus.md): In this notebook, we will use the BGE-M3 embedding model to extract embeddings from a news headline dataset, utilize Milvus to efficiently calculate distances… - [Vector Visualization](https://milvus.io/docs/vector_visualization.md): In this example, we will show how to visualize the embeddings(vectors) in Milvus using t-SNE. - [Movie Recommendation](https://milvus.io/docs/movie_recommendation_with_milvus.md): In this notebook, we will explore how to generate embeddings of movie descriptions using OpenAI and leverage those embeddings within Milvus to recommend movies… - [Funnel Search with Matryoshka Embeddings](https://milvus.io/docs/funnel_search_with_matryoshka.md): In this notebook, we examine how to use Matryoshka embeddings with Milvus for semantic search. - [Quickstart with Attu](https://milvus.io/docs/quickstart_with_attu.md): Attu is an all-in-one, open-source administration tool for Milvus. - [Use AsyncMilvusClient with asyncio](https://milvus.io/docs/use-async-milvus-client-with-asyncio.md): AsyncMilvusClient is an asynchronous MilvusClient that offers a coroutine-based API for non-blocking access to Milvus via asyncio. - [Text-to-Image Search with Milvus](https://milvus.io/docs/text_image_search.md): In this tutorial, we will explore how to implement text-based image retrieval using OpenAI’s CLIP (Contrastive Language-Image Pretraining) model and Milvus. - [Generating Milvus Query Filter Expressions with Large Language Models](https://milvus.io/docs/generating_milvus_query_filter_expressions.md): In this tutorial, we will demonstrate how to use Large Language Models (LLMs) to automatically generate Milvus filter expressions from natural language queries. ### Explore More - [Question Answering System](https://milvus.io/docs/question_answering_system.md): Build a question answering system with Milvus. - [Recommender System](https://milvus.io/docs/recommendation_system.md): Build a personalized recommender system with Milvus. - [Video Similarity Search](https://milvus.io/docs/video_similarity_search.md): Build a video similarity search system with Milvus. - [Audio Similarity Search](https://milvus.io/docs/audio_similarity_search.md): Build an audio similarity search system with Milvus. - [DNA Sequence Classification](https://milvus.io/docs/dna_sequence_classification.md): Build a DNA sequence classification system with Milvus. - [Text Search Engine](https://milvus.io/docs/text_search_engine.md): Build a text search engine with Milvus. - [Image Deduplication System](https://milvus.io/docs/image_deduplication_system.md): Build an image deduplication system with Milvus. ## FAQs - [Performance FAQs](https://milvus.io/docs/performance_faq.md): Find answers to frequently asked questions about search performance, performance enhancements, and other performance related issues. - [Product FAQs](https://milvus.io/docs/product_faq.md): Find answers to frequently asked questions about the world's most advanced vector database. - [Operational FAQs](https://milvus.io/docs/operational_faq.md): Find answers to commonly asked questions about operations in Milvus. - [Milvus Limits](https://milvus.io/docs/limitations.md): Learn about the limits while using Milvus. - [Troubleshooting](https://milvus.io/docs/troubleshooting.md): Learn about common issues you may encounter with Milvus and how to overcome them. ## API Reference Each SDK is versioned independently of the docs — the newest published version differs per SDK, so do not copy a version segment between them. - [PyMilvus (Python) — v3.0.x](https://milvus.io/api-reference/pymilvus/v3.0.x/About.md): Python SDK API reference. - [Java SDK — v3.0.x](https://milvus.io/api-reference/java/v3.0.x/About.md): Java SDK API reference. - [Go SDK — v3.0.x](https://milvus.io/api-reference/go/v3.0.x/About.md): Go SDK API reference. - [Node.js SDK — v3.0.x](https://milvus.io/api-reference/node/v3.0.x/About.md): Node.js SDK API reference. - [RESTful API — v3.0.x](https://milvus.io/api-reference/restful/v3.0.x/About.md): Language-agnostic HTTP API reference. - [C# SDK — v2.6.x](https://milvus.io/api-reference/csharp/v2.6.x/About.md): C# SDK API reference. - [C++ SDK — v3.0.x](https://milvus.io/api-reference/cpp/v3.0.x/About.md): C++ SDK API reference. - [Rust SDK — v3.0.x](https://milvus.io/api-reference/rust/v3.0.x/About.md): Rust SDK API reference. ## Interactive Learning Browser-based visualizations of how the index and metric choices actually behave. - [Learn Milvus](https://milvus.io/learn-milvus): Index of the interactive index and metric visualizations. - [IVF Visualized](https://milvus.io/learn-milvus/ivf): FLAT vs IVF search animation with a live nprobe/recall tradeoff. - [HNSW Visualized](https://milvus.io/learn-milvus/hnsw): Walk through the HNSW layered graph with tunable M and ef. - [DiskANN Visualized](https://milvus.io/learn-milvus/diskann): DiskANN's PQ-in-RAM beam search and SSD re-ranking, animated. - [Similarity Metrics Visualized](https://milvus.io/learn-milvus/metric): How L2, cosine, and inner product change which neighbors rank as nearest. ## Sizing Tools - [Milvus Sizing Tool](https://milvus.io/tools/sizing): Estimate memory, disk, and node counts for a deployment. - [Milvus Sizing Tool (Enterprise)](https://milvus.io/tools/sizing-enterprise): Sizing estimates for enterprise deployments. - [Milvus Sizing Tool (v2.5.0)](https://milvus.io/tools/sizing-v250): Sizing calculator pinned to the v2.5.0 cost model. ## External Tools Maintained outside milvus.io; these are the canonical repositories. - [Attu (Milvus GUI)](https://github.com/zilliztech/attu): Desktop/web GUI for browsing collections and running queries. - [Milvus VTS](https://github.com/zilliztech/vts): Vector Transmission Service for migrating data between collections. ## Other Site Sections These areas hold thousands of pages and are not enumerated here. Do not guess a slug inside them — resolve one through https://milvus.io/sitemap.xml or the section index below. - [Blog](https://milvus.io/blog): Release announcements, engineering deep dives, and benchmarks. Article URLs are `/blog/` with no `.md` suffix. - [AI Quick Reference](https://milvus.io/ai-quick-reference): Core concepts for vector search, indexing, RAG, data preparation, and production operations, with implementation guides and sources. - [Learn AI and Vector DB](https://milvus.io/learn-ai-and-vectordb): Long-form explainers on vector databases and AI infrastructure. - [Use Cases](https://milvus.io/use-cases): Production use cases and customer stories. - [Milvus Demos](https://milvus.io/milvus-demos): Live demo applications, including reverse image search. - [Bootcamp](https://milvus.io/bootcamp): Hands-on notebooks and end-to-end example projects. - [Community](https://milvus.io/community): Contributing, governance, and community channels.