From 93beea904b5b2c9ad1288daaa1d94b9426c52679 Mon Sep 17 00:00:00 2001 From: "Jinjing.Zhou" Date: Thu, 14 Mar 2024 17:36:13 +0800 Subject: [PATCH 1/2] Update pgvecto.rs README.md with new features and documentation link Signed-off-by: Jinjing.Zhou --- README.md | 26 +++++++++++++++++--------- 1 file changed, 17 insertions(+), 9 deletions(-) diff --git a/README.md b/README.md index 6a1abd1a6..562cb33ac 100644 --- a/README.md +++ b/README.md @@ -9,18 +9,26 @@ all-contributors

-pgvecto.rs is a Postgres extension that provides vector similarity search functions. It is written in Rust and based on [pgrx](https://github.com/tcdi/pgrx). It is currently in the beta status, we invite you to try it out in production and provide us with feedback. Read more at [📝our blog](https://blog.pgvecto.rs/pgvectors-02-unifying-relational-queries-and-vector-search-in-postgresql). +pgvecto.rs is a Postgres extension that provides vector similarity search functions. It is written in Rust and based on [pgrx](https://github.com/tcdi/pgrx). Read more at [📝our blog](https://blog.pgvecto.rs/pgvectors-02-unifying-relational-queries-and-vector-search-in-postgresql). ## Why use pgvecto.rs -- 💃 **User-Friendly**: Effortlessly incorporate pgvecto.rs into your existing database as a Postgres extension, streamlining integration with your current workflows and applications. -- 🥅 **Join and Filter without Limitation**: Elevate your search capabilities in pgvecto.rs with VBASE filtering. Apply any filter conditions and join with other tables, achieving high recall and low latency, a distinctive edge over other vector databases. -- 🌓 **Efficient FP16 Support**: Optimize your data storage with pgvecto.rs, supporting FP16 vector type to cut memory and storage usage by half, and boosting throughput. -- 🧮 **Advanced Quantization**: Utilize scalar and product quantization in pgvecto.rs for up to 64x compression. Achieve up to 4x memory savings with less than 2% recall loss with scalar quantization. -- 🔍 **Hybrid Search**: Leverage the full-text search functionality in PostgreSQL with pgvecto.rs to search text and vector data within a single query. -- 🔗 **Async indexing**: The pgvecto.rs index is built asynchronously by background threads, allowing non-blocking inserts and always ready for new queries. -- ⬆️ **Extended Vector Length**: pgvecto.rs supports vector length up to 65535, suitable for the latest models. -- 🦀 **Rust-Powered Reliability**: Rust's strict compile-time checks ensure memory safety, reducing the risk of bugs and security issues commonly associated with C extensions. +| Feature Category | Feature | | +| -------------------------- | ------------------------- | ------------------------------------------------------------------------------------------ | +| **Search Capabilities** | 🔍 Vector Search | Ultra-low-latency, high-precision vector search. | +| | 🧩 Sparse Vector Search | Keyword-based vector search using SPLADE or BM25 algorithms. | +| | 📄 Full-Text Search | Comprehensive text search across any language, powered by tsvector. | +| **Data Handling** | ✔ Complete SQL Support | Full SQL support, enabling joins and filters without limitations or extra configuration. | +| | 🔗 Async indexing | Non-blocking inserts with up-to-date query readiness. | +| | 🔄 Easy Data Management | No need for syncing vectors and metadata with external vector DB, simplifying development. | +| **Data Types** | 🔢 FP16/INT8 Data type | Supports FP16 and INT8 data types for improved storage and computational efficiency. | +| | 🌓 Binary vector support | Vector indexing with binary vectors, and Jaccard distance support. | +| | 🔪 Matryoshka embeddings | Subvector indexing, like vector[0:256], for enhanced Matryoshka embeddings. | +| | ⬆️ Extended Vector Length | Vector lengths up to 65535 supported, ideal for the latest cutting-edge models. | +| **System Performance** | 🚀 Production Ready | Battle-tested database ecosystem integrated with PostgreSQL. | +| | ⚙️ High Availability | Logical replication support to ensure high availbility. | +| | 💡 Resource Efficient | Efficient attribute storage leveraging PostgreSQL. | +| **Security & Permissions** | 🔒 Permission Control | Easy access control like read-only roles, powered by PostgreSQL. | ## [Documentation](https://docs.pgvecto.rs/getting-started/overview.html) From 884f12c10f9d8c651adcb801e66e027ef1a8631f Mon Sep 17 00:00:00 2001 From: "Jinjing.Zhou" Date: Thu, 14 Mar 2024 17:38:10 +0800 Subject: [PATCH 2/2] Update pgvecto-rs Docker image version to pg16-v0.2.1 Signed-off-by: Jinjing.Zhou --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index 562cb33ac..93e78166d 100644 --- a/README.md +++ b/README.md @@ -53,7 +53,7 @@ docker run \ --name pgvecto-rs-demo \ -e POSTGRES_PASSWORD=mysecretpassword \ -p 5432:5432 \ - -d tensorchord/pgvecto-rs:pg16-v0.2.0 + -d tensorchord/pgvecto-rs:pg16-v0.2.1 ``` Then you can connect to the database using the `psql` command line tool. The default username is `postgres`, and the default password is `mysecretpassword`.