The SQL Vector
Database for Scalable AI

MYSCALE

Enable every developer to build production-grade GenAI applications with powerful and familiar SQL. Minimal Learning, Max Value, and Cost-Effective.

Trusted SQL/Integrated Vector Database

Fully SQL-Compatible Vector Database

Fast and Powerful Queries

Perform fast, powerful, and efficient vector search, text search, filtered search, and SQL-vector join queries.

Vector search

Text search

Filtered search

Complex queries

  
    
-- Create MSTG index on vector column for arxiv_table
ALTER TABLE doc_table
ADD VECTOR INDEX vec_idx vector
TYPE MSTG('metric_type=Cosine');

-- Perform vector search and return top 10 results
SELECT
    id, title, text
    distance(vector, query_vector) as dist
FROM doc_table
ORDER BY
    dist ASC
LIMIT 10;

Manage All Types of Data

Data Import & Export

Integrate with Your Stack

Vector search

Text search

Filtered search

Complex queries

  
    
-- Create MSTG index on vector column for arxiv_table
ALTER TABLE doc_table
ADD VECTOR INDEX vec_idx vector
TYPE MSTG('metric_type=Cosine');

-- Perform vector search and return top 10 results
SELECT
    id, title, text
    distance(vector, query_vector) as dist
FROM doc_table
ORDER BY
    dist ASC
LIMIT 10;

Enhanced RAG Effectiveness

Achieve superior RAG (Retrieval-Augmented Generation) results
Combine vector and full-text search for thorough data retrieval
Utilize advanced metadata filtering for precise outcomes
Adapt to changing business requirements with flexible data modeling

High-Performance and Cost-Efficient

3x the speed, 3x the savings with MSTG vector engine
Performant & powerful metadata filter search
Faster data ingestion with optimized index building

SQL/Relational Vector Database

Manage and query AI-related data within a single system
Execute advanced SQL vector joint queries
Enable text-to-SQL translation for complex querying needs

Powerful Full-Text Search

BM25 for enhanced relevance ranking
Configurable tokenizers for various languages
Simple yet powerful query language

Integrates with Your AI Stack

Deep integration with popular development languages and frameworks
Seamless LLM application observability with MyScale Telemetry

Security and Compliance

SQL-based RBAC

In compliance with SOC 2

ISO 27001 Compliance

The Generative AI Stack

Build and deliver quickly with the best integrations and tools

Work seamlessly with tools and popular languages adopted by leading Generative AI enterprises and LLM developers

AI integrations

Language SDK

File formats

Functions

Build Your AI Apps with MyScale

RAG for Domain Knowledge Base

Recommendations

Chatbot

Image Search

Why Industry Leaders Choose MyScale

MyScale's support for SQL is just amazing! It enables developers who are familiar with SQL but not vector search to quickly get started. You won't have this kind of user experience in other products. Additionally, for production-level applications, MyScale has significantly reduced our costs.

We have previously tried vector solutions like Pinecone, PgVector, Chroma, Zilliz, and Redis. We are very impressed by the user-friendliness of MyScale and its seamless integration with Langchain.

MyScale's performance in memory resource optimization and data insertion efficiency far surpasses our previous solution.

Vector data processing is a critical part for LLM infrastructure, while SQL can provide significant scalability and convenience to application developers. MyScale has made notable contributions in this area. With its open source availability, developers will have options to leverage the value of structured data to work with different LLMs for diversified use cases, for better performance, lower cost, and faster innovation paces.

The self-developed MSTG index by MyScale has helped us resolve many issues. Additionally, in the application of Text to SQL integration, MyScale has demonstrated essential compatibility. We are truly grateful for the assistance and suggestions provided by MyScale.