Hugging Face's logo
Activity Feed Request to join this org

AI & ML interests

None defined yet.
cassandrapgml 
posted an update over 1 year ago
view post
Post
834
Korvus: Single-query RAG with Postgres

Hey folks!

We built Korvus, an open-source RAG (Retrieval-Augmented Generation) pipeline that consolidates the entire RAG workflow - from embedding generation to text generation - into a single SQL query, significantly reducing architectural complexity and latency.

https://github.com/postgresml/korvus

Here's some of the highlights:

- Full RAG pipeline (embedding generation, vector search, reranking, and text generation) in one SQL query
- SDKs for Python, JavaScript, and Rust (more languages planned)
- Built on PostgreSQL, leveraging pgvector and pgml
- Open-source, with support for open models
- Designed for high performance and scalability

We're eager to get feedback from the community and welcome contributions. Check out our GitHub repo for more details:

https://github.com/postgresml/korvus
  • 1 reply
·
cassandrapgml 
posted an update over 1 year ago
view post
Post
735
Looking to move away from OpenAI's closed-source models?

We've made switching from to open-source as easy as possible with a drop-in replacement OpenAI’s chat completion endpoint. You can specify any model you'd like in just a few lines of code. We call it the OpenAI Switch Kit.

Check out the docs here: https://postgresml.org/docs/guides/opensourceai

Or the blog post here: https://postgresml.org/blog/introducing-the-openai-switch-kit-move-from-closed-to-open-source-ai-in-minutes

Happy building!
cassandrapgml 
posted an update over 1 year ago
view post
Post
445
How to generate LLM embeddings with open source models from Hugging Face 🤗 in PostgresML.

This article is the first in a multipart series that will show you how to build a post-modern semantic search and recommendation engine.

➡️ https://postgresml.org/blog/generating-llm-embeddings-with-open-source-models-in-postgresml

PostgresML is a backend for your AI app that unifies LLMs w/ vector memory + embedding generation + reranking & pruning models — all in a single process for better performance.

We're always looking for ways to make PostgresML better — let us know what you think!