description:What is PostgresML?

coverY:0

layout:

cover:

visible:false

size:full

title:

visible:true

description:

visible:true

tableOfContents:

visible:true

outline:

visible:true

pagination:

visible:true

PostgresML supercharges your Postgres database into an end-to-end MLOps platform, seamlessly integrating the key components of the machine learning workflow. Without moving data outside your database, PostgresML allows Postgres to function as a feature store, model store, training engine, and inference service all in one place. This consolidation streamlines building and deploying performant, real-time AI applications for developers.

[PostgresML](https://postgresml.org/) is a PostgreSQL extension that enables users to build fast, simple and powerful models right inside your database. Users can build chatbots, search engines, forecasting apps and more with the latest NLP, LLM and ML models using the simplicity and power of SQL (and our SDKs). 

Today's engineers and product teams are increasingly burdened with data hungry ML & AI apps and complex architecture. Several roundtrip network calls must be made across a web of microservices to return an accurate response. This approach to MLops is slower for users, and it’s more complex to build, manage and scale. 

PostgresML collocates data and compute, so you can save models and index data right in your Postgres database. When you need to predict or train using machine learning, you don’t have to make requests over the internet or even take your data out of your database - which is both slower and a data safety risk. We’ve even added GPUs to our databases to fully leverage the latest technology and algorithms. You can avoid the complexity and latency of microservices when you have a complete MLops platform right in your database.

Instead of patching together MongoDB, Databricks, Pinecone, Huggingface, LangChain and more to get the latest algorithms and LLMs in your app – you can add one extension to your Postgres database and get to market quickly. 


Why Postgres? If it’s not broke, don’t fix it. Postgres is old reliable for a reason. It’s highly efficient, scalable and open-source (Just like all of PostgresML). Plus, you can easily build your AI app using simple SQL.

In ML applications, new data is constantly being generated by users and needs to be stored somewhere (feature store). For example, there are long term historical features, short term session level, and real time request level. Postgres can handle both long term with partitioning, table spaces and indexing as well as short term storage, and accept real-time features as parameters in queries, or as additional session level storage already written during the request.

In database-ML minimizes latency and computational cost. Postgres can also handle all types of data efficiently; including vectors, geospatial, JSON, timeseries, tabular and text. Our benchmarks show up to**40x faster** improvement over Python microservices.

* Torch

* Tensorflow

* SCikit Learn

* XGBoost

* LightGBM

* Pre-trained deep learning models from Hugging Face

* LLama 2

* Falcon LLM

* OpenAI

With PostgresML, your database becomes a full-fledged ML workbench. It supports supervised and unsupervised algorithms like regression, clustering, deep neural networks,andmore. You can build models using SQL on data insidePostgres. Models are stored back into Postgres for low-latency inferences later.

We’re a totally open-source project. That includes various ML librariesandthe vastPostgres ecosystem.

PostgresML isopen source but also offered as a fully-managed cloud service. In addition to the SQL API, it provides Javascript, Python, and Rust SDKs to quicklybuildvector search, chatbots,andother ML appsin just a few linesofcode.

[PostgresML](https://postgresml.org/) isin early stages of development - but we are moving quickly. It’s our vision to help as many app developers as possiblebuild andscale AI appswithout allofthe complexity.

To scale horizontally, PostgresML utilizes PgCat, an advanced PostgreSQL proxy and load balancer. PgCat enables sharding, load balancing, failover, and mirroringtoachieve extremely high throughput and low latency. By keepingtheentire machine learning workflow within Postgres, PostgresML avoids expensive network calls between disparate systems. This allows PostgresML to handle millions of requests per second at up to 40x the speed of other platforms. PgCat and Postgres replication deliver seamless scaling while retaining transactional integrity.

If you’d like to stay up-to-date with our progress or contributeto theproject (feedback, questions and comments are all welcome) you can check us out on[Github](https://github.com/postgresml/postgresml) or chat with us anytime on our[Discord](https://discord.com/invite/DmyJP3qJ7U).

-v postgresml_data:/var/lib/postgresql \

-p 5433:5432 \

-p 8000:8000 \

ghcr.io/postgresml/postgresml:2.7.12 \

ghcr.io/postgresml/postgresml:2.7.3 \

sudo -u postgresml psql -d postgresml

--gpus all \

-p 5433:5432 \

-p 8000:8000 \

ghcr.io/postgresml/postgresml:2.7.12 \

ghcr.io/postgresml/postgresml:2.7.3 \

sudo -u postgresml psql -d postgresml