PostgresML | You know Postgres. Now you know machine learning – PostgresML (original) (raw)
Postgres with
GPUs for
ML/AI apps
Index, filter & rank vectors
Generate real-time, fact-based outputs
[{“translation_text”:”Bienvenue à l'avenir!”}]AI is going to change the world!
"Bleeding edge stuff in
a
matter of minutes."
Stuck with an AI stack so complicated your app barely runs in prod?🤔
handyman
Microservice mayhem
You're managing a multitude of microservices - a vector database, embedding model, LLMs, and frameworks to glue them all together.
cognition
Increasing inefficiency
Production outages that won't stop, high-latency UX, ever-increasing dev time, and data-hungry compute with costly vendors.
mystery
Excessive exposure
Your data is sent through multiple systems. You can't be sure if it's secure, stable, compliant or private.
Architecture makes or breaks your app.
PostgresML radically simplifies it
"Over the past year, the data infrastructure stack has seen substantial stability in core systems and rapid proliferation of supporting tools and applications" - a16z
4x Faster
than 
HuggingFace + 
Pinecone
for a RAG chatbot
10x faster
than 
OpenAI for embedding
generation
Save 42%
On vector database cost
compared to Pinecone
Don't take our word for it.
Explore the SDK and test open source models in our hosted database.
What makes PostgresML so powerful
Index, filter and re-rank vector embeddings
10x faster vector operations
Perform fast KNN and ANN search
Index embeddings with HNSW or IVFFlat
Generate embeddings
Choose from state-of-the-art models
Built-in data preprocessors for splitting and chunking
Convert text to vector embeddings
Colocate data and compute
Embed, serve and store all in one process
Terabytes of data on a single machine
Built-in data privacy & security
Train, tune and deploy
Regression, classification and clustering
Fine-tune LLMs on your own data
Monitor model deployments over time
Get the most of LLMs
Use open-source models (Mistral, LLama, etc.)
Perform a range of NLP tasks
Serve with the same infrastructure
Comprehensive platform
Multiple deployment options
Perform several AI & machine learning tasks
Use SQL or SDKs in JS and Python
Index, filter and re-rank vector embeddings
10x faster vector operations
Perform fast KNN and ANN search
Index embeddings with HNSW or IVFFlat
Generate embeddings
Choose from state-of-the-art models
Built-in data preprocessors for splitting and chunking
Convert text to vector embeddings
Colocate data and compute
Embed, serve and store all in one process
Terabytes of data on a single machine
Built-in data privacy & security
Train, tune and deploy
Regression, classification and clustering
Fine-tune LLMs on your own data
Monitor model deployments over time
Get the most of LLMs
Use open-source models (Mistral, LLama, etc.)
Perform a range of NLP tasks
Serve with the same infrastructure
Comprehensive platform
Multiple deployment options
Perform several AI & machine learning tasks
Use SQL or SDKs in JS and Python
Better price for performance
Our pricing is based on the models you use. It’s designed to minimize costs and operations. You’ll also save because you can replace many existing tools.
Integrated Libraries
add remove
PyTorch
TensorFlow
Flax
SciKit-Learn
Hugging Face
Llama
Mistral
XGBoost
LightGBM
CatBoost
Mixtral
Mistral
dbrx-instruct
Haskell
Java & Scala
Julia
Lua
Apache Airflow
DBT
DBeaver
Azure
Google Cloud
Work with
what you want
Hear from our community
This is why I’m bullish on @postgresml - devs will always prefer to do things in data stores they already use in production
Great article by PostgresML, running @huggingface models INSIDE @PostgreSQL nice tidbit on scalability: "Our example data is based on 5 million DVD reviews from Amazon ... that's more data than fits in a Pinecone Pod at the time of writing"
Love the fact that @postgresml can run various algorithms to find the optimum one for model creation
You can look at PostgresML. Its based on Postgres, not specifically a vector database but they've got a pleasantly full featured eco-system for the whole training process, fetching datasets, huggingface integration, training etc. of course they also have vector related functions
If you want to seamlessly integrate machine learning models into your #PostgreSQL database, use PostgresML.
Khuyen Tran
@KhuyenTran16
💯 there's also PostgresML if you wanna get a little more full featured - supports embedding in-database as well as CUBE / pgvector
Tons of capability in that Postgres extension. It's an important part of the ML Stack at cloud.tembo.io as well.
A game-changer indeed! By integrating ML and AI directly at the database level with @postgresml, we're not just streamlining processes but revolutionizing data handling and insights generation in one fell swoop.
This is why I’m bullish on @postgresml - devs will always prefer to do things in data stores they already use in production
Great article by PostgresML, running @huggingface models INSIDE @PostgreSQL nice tidbit on scalability: "Our example data is based on 5 million DVD reviews from Amazon ... that's more data than fits in a Pinecone Pod at the time of writing"
Love the fact that @postgresml can run various algorithms to find the optimum one for model creation
You can look at PostgresML. Its based on Postgres, not specifically a vector database but they've got a pleasantly full featured eco-system for the whole training process, fetching datasets, huggingface integration, training etc. of course they also have vector related functions
If you want to seamlessly integrate machine learning models into your #PostgreSQL database, use PostgresML.
Khuyen Tran
@KhuyenTran16
💯 there's also PostgresML if you wanna get a little more full featured - supports embedding in-database as well as CUBE / pgvector
Tons of capability in that Postgres extension. It's an important part of the ML Stack at cloud.tembo.io as well.
A game-changer indeed! By integrating ML and AI directly at the database level with @postgresml, we're not just streamlining processes but revolutionizing data handling and insights generation in one fell swoop.
Get started
with$100 in
free credits
