Duckle v0.4.1 is out - DuckDB 1.5.4, in-app updates, Custom SQL for duck sources, and proxy support for REST. Read the release notes

Move data from anything,
to anywhere, on your machine.

Duckle is a local-first ETL studio with a built-in AI assistant. Connect 160+ sources and destinations, build pipelines visually or describe them in plain English, and run them at native DuckDB speed. No cloud, no servers, no lock-in.

Free & open source No account, no telemetry Runs fully offline
329components, all on DuckDB
104 / 59sources / destinations
5M rowsloaded in about 1 second
100%local - data never leaves
Integrations

Move data across 160+ connectors, working today

Files, databases, warehouses, object stores, streaming brokers, SaaS APIs, NoSQL and vector DBs. These work today, not a coming-soon list.

PostgreSQLMySQLMariaDBSQLiteSQLiteSQL ServerOracleMongoDBRedisCassandraClickHouseCockroachDBNeo4jInfluxDBElasticsearchDuckDBSnowflakeBigQueryDatabricksRedshiftParquetSparkHadoop PostgreSQLMySQLMariaDBSQLiteSQLiteSQL ServerOracleMongoDBRedisCassandraClickHouseCockroachDBNeo4jInfluxDBElasticsearchDuckDBSnowflakeBigQueryDatabricksRedshiftParquetSparkHadoop
Amazon S3Google Cloud StorageAzureKafkaKafkaPulsarRabbitMQNATSSalesforceHubSpotStripeShopifyNotionNotionGitHubGitHubGitLabJiraSlackDiscordAsanaTrelloZendeskTwilioTelegramMailchimpGoogle AnalyticsGoogle SheetsAirtableQdrant Amazon S3Google Cloud StorageAzureKafkaKafkaPulsarRabbitMQNATSSalesforceHubSpotStripeShopifyNotionNotionGitHubGitHubGitLabJiraSlackDiscordAsanaTrelloZendeskTwilioTelegramMailchimpGoogle AnalyticsGoogle SheetsAirtableQdrant
View all integrations →
The visual Map editor with a main input, two lookups, typed output expressions and a filter
Build

Automate pipelines end to end, without writing SQL

Drag sources, transforms, validators and sinks onto a canvas and wire them together. The visual Map editor joins a main input to up to three lookups with per-output typed expressions and an inline filter. Every node still compiles to readable SQL you can inspect in the Plan tab.

  • 128 transforms: joins, windows, aggregates, CDC/SCD, AI, quality
  • Validators route bad rows to a reject port
  • Generated SQL on every node - never a black box
Run & monitor

Native speed, with live operational visibility

Execution runs through DuckDB - vectorized and columnar. Branches fan out across your CPU cores automatically. Every run reports per-node status, row counts, timings and a live preview, so you see exactly what happened and where.

  • 16-node, multi-branch runs finish in a few seconds
  • Run history, plans and per-node previews built in
  • Schedule on cron / interval / file-watch, or run headless
A run summary showing 16 nodes and 279 rows written in 3006 ms, with a live preview
Solutions

One studio for every data movement job

Replication

Warehouse loading & CDC

Upsert into Snowflake, BigQuery or Postgres on a schedule, with change data capture and delete propagation.

DuckLake CDC mirror pipeline
SaaS & APIs

Pull from apps & services

Salesforce, HubSpot, Stripe, GitHub, Jira and generic REST / GraphQL / OData - normalized into rows you can join.

Component palette and Duckie assistant
Files & storage

Lakes, files & object stores

Glob S3 / GCS / Azure, read Parquet, Iceberg, Delta and DuckLake, and write Hive-partitioned output at native speed.

Incremental load pipeline
Duckie AI assistant panel beside the canvas
Local AI

Describe it. Duckie builds it. On device.

The assistant runs Qwen 2.5 Coder locally through llama.cpp - no API key, no cloud. Ask in plain English; Duckie streams a valid pipeline and drops positioned, wired nodes onto the canvas. The same six AI transforms prep data for RAG, all offline.

  • Truly local - prompts and data never leave the machine
  • Chunk, redact PII, embed and dedupe for RAG
  • Connect Claude or any LLM over the built-in MCP server
Proof, not promises

Real pipelines, real run results

Connecting the dots in the DuckDB ecosystem

Duckle turns DuckDB into a visual ETL platform

Every pipeline compiles to SQL and runs through DuckDB. Duckle sits in the ingestion and transformation layer alongside tools like dlt, dbt and Evidence, and reads the wider family natively: DuckLake tables, MotherDuck, the Quack protocol, and the community extensions.

Duckle is an independent, open-source project by SlothFlowLabs. It builds on the DuckDB engine but is not part of, affiliated with, or endorsed by DuckDB Labs or MotherDuck.
Get started

Build your first pipeline in 60 seconds

Free and open source, dual-licensed MIT OR Apache-2.0. Windows, macOS and Linux. No account, no telemetry.