# ParadeDB - [Welcome to ParadeDB](https://www.paradedb.com/docs/start/introduction.md): Just use Postgres - [Install ParadeDB](https://www.paradedb.com/docs/start/install.md): Run ParadeDB locally - [Configure Your Environment](https://www.paradedb.com/docs/start/configure-your-environment.md): Connect your preferred SQL client or ORM to ParadeDB - [Create Your First Index](https://www.paradedb.com/docs/start/create-your-first-index.md): Create a ParadeDB index over your table - [Run Your First Queries](https://www.paradedb.com/docs/start/run-queries.md): Run your first queries on ParadeDB - [Load Your Own Data](https://www.paradedb.com/docs/start/load-data.md): Dump data from an existing Postgres and load into ParadeDB - [Set Up Your Coding Agent](https://www.paradedb.com/docs/start/ai-agents.md): Teach your coding agent to use ParadeDB - [Go to Production](https://www.paradedb.com/docs/start/go-to-production.md): Move from the sandbox to a production deployment - [Reference](https://www.paradedb.com/docs/reference/overview.md): Detailed syntax for ParadeDB indexes, queries, aggregates, and SQL APIs - [Create an Index](https://www.paradedb.com/docs/reference/indexing/create-index.md): Index a Postgres table for ParadeDB search workloads - [Partial Indexes](https://www.paradedb.com/docs/reference/indexing/indexing-partial.md): Add row filters to the ParadeDB index - [Columnar Storage](https://www.paradedb.com/docs/reference/indexing/columnar.md): Column-oriented indexing for fast filtering, sorting, and aggregates - [Index Partitioning](https://www.paradedb.com/docs/reference/indexing/partition-by.md): Partition ParadeDB indexes to accelerate filter and JOIN queries - [Faster BM25 Queries](https://www.paradedb.com/docs/reference/indexing/faster-bm25-queries.md): Speed up queries that return the most relevant results - [Indexing Expressions](https://www.paradedb.com/docs/reference/indexing/indexing-expressions.md): Add Postgres expressions to the index - [Indexing JSON](https://www.paradedb.com/docs/reference/indexing/indexing-json.md): Add JSON and JSONB types to the index - [Indexing Text Arrays](https://www.paradedb.com/docs/reference/indexing/indexing-arrays.md): Add text arrays to the index - [Indexing 32+ Columns](https://www.paradedb.com/docs/reference/indexing/indexing-composite.md): Use composite types to index more than 32 columns - [Indexing Vectors](https://www.paradedb.com/docs/reference/indexing/indexing-vectors.md): Vectors live alongside text and filters in the ParadeDB index - [Full-Text / Lexical Search](https://www.paradedb.com/docs/reference/full-text/overview.md): Full-text query syntax, scoring, filtering, sorting, and highlighting - [Tokenizers](https://www.paradedb.com/docs/reference/tokenizers/overview.md): Tokenizers split large chunks of text into small, searchable units called tokens - [Unicode](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/unicode.md): The default text tokenizer in ParadeDB - [Literal](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/literal.md): Indexes the text in its raw form, without any splitting or processing - [Literal Normalized](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/literal-normalized.md): Like the literal tokenizer, but allows for token filters - [Whitespace](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/whitespace.md): Tokenizes text by splitting on whitespace - [Ngram](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/ngrams.md): Splits text into small chunks called grams, useful for partial matching - [Edge Ngram](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/edge-ngrams.md): Generates prefix n-grams per word, ideal for search-as-you-type - [Simple](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/simple.md): Splits on any non-alphanumeric character - [Regex Patterns](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/regex.md): Tokenizes text using a regular expression - [Chinese Compatible](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/chinese-compatible.md): A simple tokenizer for Chinese, Japanese, and Korean characters - [Lindera](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/lindera.md): Uses prebuilt dictionaries to tokenize Chinese, Japanese, and Korean text - [ICU](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/icu.md): Splits text according to the Unicode standard - [Jieba](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/jieba.md): The most advanced Chinese tokenizer that leverages both a dictionary and statistical models - [Source Code](https://www.paradedb.com/docs/reference/tokenizers/available-tokenizers/source-code.md): Tokenizes text that is actually code - [Token Filters](https://www.paradedb.com/docs/reference/token-filters/overview.md): Configure lowercasing, stemming, ASCII folding, and other token processing - [Alpha Numeric Only](https://www.paradedb.com/docs/reference/token-filters/alphanumeric.md): Removes any tokens that contain characters that are not ASCII letters or digits - [ASCII Folding](https://www.paradedb.com/docs/reference/token-filters/ascii-folding.md): Strips away diacritical marks like accents - [Lowercase](https://www.paradedb.com/docs/reference/token-filters/lowercase.md): Converts all characters to lowercase - [Remove Stopwords](https://www.paradedb.com/docs/reference/token-filters/stopwords.md): Remove language-specific stopwords from the index - [Stemmer](https://www.paradedb.com/docs/reference/token-filters/stemming.md): Reduces words to their root form for a given language - [Token Length](https://www.paradedb.com/docs/reference/token-filters/token-length.md): Remove tokens that are above or below a certain byte length from the index - [Trim](https://www.paradedb.com/docs/reference/token-filters/trim.md): Remove trailing and leading whitespace from a token - [Multiple Tokenizers Per Field](https://www.paradedb.com/docs/reference/tokenizers/multiple-per-field.md): Apply different token configurations to the same field - [Query-Time Tokenizer](https://www.paradedb.com/docs/reference/tokenizers/search-tokenizer.md): Use a different tokenizer at search time than at index time - [Advanced Query Functions](https://www.paradedb.com/docs/reference/full-text/query-builder.md): Query builder functions for advanced full-text query types - [Match](https://www.paradedb.com/docs/reference/full-text/match.md): Returns documents that match the provided query string, which is tokenized before matching - [Term](https://www.paradedb.com/docs/reference/full-text/term.md): Look for exact token matches in the source document, without any further processing of the query string - [Phrase](https://www.paradedb.com/docs/reference/full-text/phrase.md): Phrase queries are like match queries, but with order and position of matching tokens enforced - [Proximity](https://www.paradedb.com/docs/reference/full-text/proximity.md): Match documents based on token proximity within the source document - [Phrase Prefix](https://www.paradedb.com/docs/reference/full-text/phrase-prefix.md): Finds documents containing a phrase followed by a term prefix - [Regex Phrase](https://www.paradedb.com/docs/reference/full-text/regex-phrase.md): Matches a specific sequence of regex queries - [Fuzzy](https://www.paradedb.com/docs/reference/full-text/fuzzy.md): Allow for typos in the query string - [Regex](https://www.paradedb.com/docs/reference/full-text/regex.md): Searches for terms that match a regex pattern - [Range Term](https://www.paradedb.com/docs/reference/full-text/range-term.md): Filters over Postgres range types - [More Like This](https://www.paradedb.com/docs/reference/full-text/more-like-this.md): Finds documents that are "like" another document - [Query Parser](https://www.paradedb.com/docs/reference/full-text/query-parser.md): Accept raw user-provided query strings - [All](https://www.paradedb.com/docs/reference/full-text/all.md): Search all rows in the index - [Overview](https://www.paradedb.com/docs/reference/filtering/overview.md): Choose the right filtering path for ParadeDB queries - [Indexed Columns](https://www.paradedb.com/docs/reference/filtering/indexed.md): Push filters into the ParadeDB index scan - [External Indexes](https://www.paradedb.com/docs/reference/filtering/external-indexes.md): Combine ParadeDB search with filters backed by other database indexes - [Top K](https://www.paradedb.com/docs/reference/full-text/top-k.md): ParadeDB is optimized for quickly finding the Top K results in a table - [BM25 Scoring](https://www.paradedb.com/docs/reference/full-text/score.md): BM25 scores sort the result set by relevance - [Relevance Tuning](https://www.paradedb.com/docs/reference/full-text/relevance-tuning.md): Tune the BM25 score by adjusting the weights of individual queries - [Highlighting](https://www.paradedb.com/docs/reference/full-text/highlight.md): Generate snippets for portions of the source text that match the query string - [Vector / Semantic Search](https://www.paradedb.com/docs/reference/vector/overview.md): Reference pages for indexing, querying, and tuning vector and semantic search - [Querying Vectors](https://www.paradedb.com/docs/reference/vector/querying.md): Run nearest-neighbor vector search inside the ParadeDB index - [Tuning Recall and Latency](https://www.paradedb.com/docs/reference/vector/tuning.md): Trade recall against latency for vector search - [Hybrid Search](https://www.paradedb.com/docs/reference/hybrid/overview.md): Reference pages for combining full-text and vector rankings - [Reciprocal Rank Fusion](https://www.paradedb.com/docs/reference/hybrid/rrf.md): Combine full text and vector search results into a single ranking - [Aggregates](https://www.paradedb.com/docs/reference/aggregates/overview.md): Accelerate SQL and JSON aggregates with the ParadeDB index - [Terms](https://www.paradedb.com/docs/reference/aggregates/bucket/terms.md): Count the number of occurrences for each value in a result set - [Filters](https://www.paradedb.com/docs/reference/aggregates/bucket/filters.md): Compute aggregations over multiple filters in one query - [Histogram](https://www.paradedb.com/docs/reference/aggregates/bucket/histogram.md): Count the number of occurrences over some interval - [Date Histogram](https://www.paradedb.com/docs/reference/aggregates/bucket/datehistogram.md): Count the number of occurrences over fixed time intervals - [Range](https://www.paradedb.com/docs/reference/aggregates/bucket/range.md): Count the number of occurrences over user-defined buckets - [Average](https://www.paradedb.com/docs/reference/aggregates/metrics/average.md): Compute the average value of a field - [Cardinality](https://www.paradedb.com/docs/reference/aggregates/metrics/cardinality.md): Compute the number of distinct values in a field - [Count](https://www.paradedb.com/docs/reference/aggregates/metrics/count.md): Count the number of values in a field - [Min/Max](https://www.paradedb.com/docs/reference/aggregates/metrics/minmax.md): Compute the min/max value of a field - [Percentiles](https://www.paradedb.com/docs/reference/aggregates/metrics/percentiles.md): Analyze the distribution of a field - [Stats](https://www.paradedb.com/docs/reference/aggregates/metrics/stats.md): Compute several metrics at once - [Sum](https://www.paradedb.com/docs/reference/aggregates/metrics/sum.md): Compute the sum of a field - [Top Hits](https://www.paradedb.com/docs/reference/aggregates/metrics/tophits.md): Compute the top hits for each bucket in a terms aggregation - [Facets](https://www.paradedb.com/docs/reference/aggregates/facets.md): Compute a Top K and aggregate in one query - [Limitations](https://www.paradedb.com/docs/reference/aggregates/limitations.md): Caveats for aggregate support - [Supported Join Types](https://www.paradedb.com/docs/reference/joins/overview.md): JOIN types and pushdown requirements supported by ParadeDB - [Using ParadeDB with Citus](https://www.paradedb.com/docs/reference/sharding/citus.md): Distributed full-text search with Citus and ParadeDB - [Operator Reference](https://www.paradedb.com/docs/reference/operators-and-functions.md): ParadeDB operators and their SQL function equivalents - [Configuration Reference](https://www.paradedb.com/docs/reference/configuration.md): Runtime settings for ParadeDB and pg_search - [SQL Function Reference](https://www.paradedb.com/docs/reference/sql-functions.md): User-facing SQL functions exposed by ParadeDB and pg_search - [Architecture](https://www.paradedb.com/docs/concepts/architecture.md): A deep dive into how ParadeDB is built - [Guarantees](https://www.paradedb.com/docs/concepts/guarantees.md): ParadeDB ensures ACID compliance, concurrency, data integrity, and replication safety - [How Text Search Works](https://www.paradedb.com/docs/concepts/full-text/overview.md): Understand how ParadeDB uses token matching to efficiently search large corpuses of text - [How Vector Search Works](https://www.paradedb.com/docs/concepts/vector/overview.md): Understand how ParadeDB natively supports vector search - [How Hybrid Search Works](https://www.paradedb.com/docs/concepts/hybrid/overview.md): Understand how ParadeDB combines text and vector search into a single ranking - [Limitations & Tradeoffs](https://www.paradedb.com/docs/concepts/limitations.md): Understand ParadeDB's key limitations and tradeoffs - [ParadeDB Enterprise](https://www.paradedb.com/docs/operate/deploy/enterprise.md): Feature comparison between ParadeDB Community and Enterprise - [Deployment Options](https://www.paradedb.com/docs/operate/deploy/overview.md): Explore the different ways to deploy ParadeDB into production - [ParadeDB Cloud](https://www.paradedb.com/docs/operate/deploy/cloud.md): Run ParadeDB on our fully managed platform - [ParadeDB BYOC](https://www.paradedb.com/docs/operate/deploy/byoc.md): Run ParadeDB in your cloud, managed by our team - [Kubernetes](https://www.paradedb.com/docs/operate/deploy/self-hosted/kubernetes.md): Run ParadeDB in an existing Kubernetes cluster - [Self-Hosted PostgreSQL](https://www.paradedb.com/docs/operate/deploy/self-hosted/extension.md): Run ParadeDB in a self-managed Postgres cluster - [Railway](https://www.paradedb.com/docs/operate/deploy/cloud-platforms/railway.md): Deploy ParadeDB on Railway - [Render](https://www.paradedb.com/docs/operate/deploy/cloud-platforms/render.md): Deploy ParadeDB on Render - [Fly.io](https://www.paradedb.com/docs/operate/deploy/cloud-platforms/fly.md): Deploy ParadeDB on Fly.io - [Dokku](https://www.paradedb.com/docs/operate/deploy/cloud-platforms/dokku.md): Deploy ParadeDB on Dokku - [DigitalOcean](https://www.paradedb.com/docs/operate/deploy/cloud-platforms/digitalocean.md): Deploy ParadeDB on DigitalOcean - [High Availability](https://www.paradedb.com/docs/operate/deploy/self-hosted/high-availability.md): Use read replicas to minimize downtime in production - [Getting Started](https://www.paradedb.com/docs/operate/deploy/logical-replication/getting-started.md): Configure ParadeDB as a logical subscriber of an existing Postgres primary - [Initial Sync and Indexing](https://www.paradedb.com/docs/operate/deploy/logical-replication/bootstrapping.md): Wait for the initial copy, build ParadeDB indexes, and query replicated data - [Replication Configuration](https://www.paradedb.com/docs/operate/deploy/logical-replication/configuration.md): Configure publication boundaries, replication capacity, and replica identity - [Schema Changes](https://www.paradedb.com/docs/operate/deploy/logical-replication/schema-changes.md): Add replicated tables and safely apply schema changes - [Monitoring and Troubleshooting](https://www.paradedb.com/docs/operate/deploy/logical-replication/monitoring-and-troubleshooting.md): Monitor replication health, resolve apply failures, and recover from WAL buildup - [Multi-Database Replication](https://www.paradedb.com/docs/operate/deploy/logical-replication/multi-database.md): Consolidate multiple microservice databases into a single ParadeDB instance for app-wide search and cross-database joins - [Reindexing](https://www.paradedb.com/docs/operate/index-maintenance/reindexing.md): Rebuild an existing index with zero downtime - [Verify Index Integrity](https://www.paradedb.com/docs/operate/index-maintenance/verify-index.md): Inspect ParadeDB indexes for corruption and structural issues - [How to Tune ParadeDB](https://www.paradedb.com/docs/operate/performance-tuning/overview.md): Tune PostgreSQL settings for better read and write performance - [Index Creation](https://www.paradedb.com/docs/operate/performance-tuning/create-index.md): Settings to make index creation faster - [Read Throughput](https://www.paradedb.com/docs/operate/performance-tuning/reads.md): Settings to improve read performance - [Write Throughput](https://www.paradedb.com/docs/operate/performance-tuning/writes.md): Settings to improve write performance - [Upgrading ParadeDB](https://www.paradedb.com/docs/operate/deploy/upgrading.md): Upgrade ParadeDB to a newer version - [GitHub Actions](https://www.paradedb.com/docs/operate/deploy/ci/github-actions.md): Run ParadeDB in GitHub Actions - [GitLab Pipelines](https://www.paradedb.com/docs/operate/deploy/ci/gitlab-pipelines.md): Run ParadeDB in GitLab Pipelines - [Roadmap](https://www.paradedb.com/docs/project/roadmap.md): The main features that we are currently working on - [Changelog](https://www.paradedb.com/docs/project/changelog.md): Release notes for ParadeDB - [0.25.11](https://www.paradedb.com/docs/project/changelog/0.25.11.md): ParadeDB release notes for 0.25.11 - [Help and Support](https://www.paradedb.com/docs/project/support.md): How to obtain support for ParadeDB This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.