Build AI apps
in Postgres

PostgresML is a complete MLOps platform in a PostgreSQL extension. Build simpler, faster and more scalable models right inside your database.

TASK:

MODEL:

[{“translation_text”:”Bienvenue à l'avenir!”}]

AI is going to change the world!

trusted by:

Trusted by:

"Bleeding edge stuff in
a
matter of minutes."

Don't take our word for it.

Explore the SDK and test open source models in our hosted database.

TASK:

MODEL:

What can you build?

The possibilities are limitless, but you could start with a few common use cases.

smart_toy

Chatbots

Combine and automate the entire workflow from embedding generation to indexing and querying for the simplest (and fastest) knowledge-based chatbot implementation.

arrow_right_alt

manage_search

Site Search

Leverage multiple types of natural language processing (NLP) and machine learning (ML) models such as vector search and personalization with embeddings to improve search results.

(coming soon)

e911_emergency

Fraud Detection

Return results and detect fraud faster with ML at the database layer. PostgresML outperforms traditional Python microservices by a factor of 8 in local tests and a factor of 40 on AWS EC2.

(coming soon)

avg_pace

Forecasting

Leverage your data with time series forecasting to garner key business insights. Build statistical and predictive models with the full power of SQL and dozens of regression algorithms.

(coming soon)

Bring ML code to your data (base)

Whatever your use case – like in the chatbot example below – PostgresML abstracts the data management overhead from the ML/AI lifecycle by enabling users to run ML/LLM models directly on a Postgres database.

Machine learning as
easy as one, two, three

It’s simple with a seamless in-database MLOps platform.

1. Train

SELECT pgml.train( 'Sales Forecasting', task => 'regression', relation_name => 'hist_sales', y_column_name => 'next_sales', algorithm => 'xgboost' );

2. Deploy

SELECT pgml.deploy( 'Sales Forecasting', strategy => 'best_score', algorithm => 'xgboost' );

3. Predict

SELECT pgml.predict( 'Sales Forecasting', ARRAY[ last_week_sales, week_of_year ] ) AS prediction FROM hist_sales ORDER BY prediction DESC;

50+
50+
50+
supported
 algorithms

40x faster
40x faster
40x faster
than python
 microservices

4k+
4k+
4k+
github stars

1m QPS
1m QPS
1m QPS
on EC2

bolt

High Efficiency

Minimize latency and computational cost with simple, high availability infrastructure that requires fewer network calls to microservices.

Learn more arrow_right_alt

database

In-database ML

Bring ML code to your data (base). Easily build ML apps with SQL, manage less complexity and enjoy built-in security with no need for unsafe data transfers.

Learn more arrow_right_alt

code

Open-source

Consume, contribute and create with our 100% open-source platform, including various ML libraries and the vast Postgres ecosystem.

Learn more arrow_right_alt

open_in_full

Instant scalability

Scale your inference layer to millions of predictions per second with the click of a button across multiple GPUs with our custom Postgres pooler, PgCat.

Learn more arrow_right_alt

Toolkits

add remove

Hugging Face

SciKit-Learn

XGBoost

LightGBM

PyTorch

TensorFlow

Models

add remove

Llama

Falcon

Bert

WizardLM

Orca

Platypus

Mistral

TheBloke

Languages

add remove

C++

Elixir

Haskell

Java & Scala

Julia

Lua

Node

Perl

PHP

Python

Ruby

Rust

Swift

IDEs

add remove

Apache Superset

DBeaver

Data Grip

Postico 2

PopSQL

Tableau

PowerBI

Jupyter

VSCode

Work with
what you want

Start your free
project today

Get Started See Pricing

Start your free
project today

Get Started See Pricing

PostgresML

Overview Chatbot Site Search (coming soon!) Fraud Detection (coming soon!) Forecasting (coming soon!)

Documentation

Blog

Pricing

About Careers Privacy Terms of Service Contact

GitHub Discord Formerly Twitter YouTube LinkedIn

Build AI apps in Postgres

"Bleeding edge stuff in a matter of minutes."

Don't take our word for it.

What can you build?

Chatbots

Site Search

Fraud Detection

Forecasting

Bring ML code to your data (base)

Machine learning as easy as one, two, three

1. Train

1. Train

2. Deploy

2. Deploy

3. Predict

3. Predict

High Efficiency

In-database ML

Open-source

Instant scalability

Toolkits

Models

Languages

IDEs

Work with what you want

Start your free project today

Start your free project today

Docs

Docs

Blog

Blog

Community

Community

PostgresML

Build AI apps
in Postgres

"Bleeding edge stuff in
a
matter of minutes."

Machine learning as
easy as one, two, three

Work with
what you want

Start your free
project today

Start your free
project today