SingleStore Kai - Vector searches & 100X faster JSON analytics

by
Experience split-second response times and elastic scalability with SingleStore. Boost analytics on your apps by 100x with the lightning-fast MongoDB API. Unlock the ability to create embeddings and perform semantic searches using MongoDB APIs on SingleStore.

Add a comment

Replies

Best
🟣 Greetings Product Hunters! 🟣 SingleStore Kai for MongoDB enables apps written for MongoDB to work against SingleStoreDB - just switch your connection string! So what is SingleStoreDB? 🤔 Here are three things you should know: 1. SingleStoreDB stores documents in an optimized data structure that combines the best of row-store and column-store 🥞 and compiles queries using state-of-the-art LLVM to optimized machine code 🔥 2. It's really, really fast at large-scale aggregations and analytics 🚀 3. SingleStoreDB supports accelerated SIMD vector operations for AI-driven text and image queries ✨ Now all of these capabilities are available to MongoDB applications with no code changes required! Try it today at Some quick FAQ: 1. How fast is it? Our benchmarks show many aggregations are 100x to 1000x faster! Some more at 2. How can developers use it? SingleStoreDB can replace or augment an existing MongoDB compatible database (this includes MongoDB, AWS DocumentDB, and Azure CosmosDB with MongoDB API). See for more details! 3. How can I test it out? Go here for a trial account and give it a try at I'm really excited to launch this - thanks for reading! Regards, Jason
It is amazing the pace of innovation happening at SingleStore.
The fact that you can do vector functions for semantic search on JSON is phenomenal. This means you can add Gen AI features with LLM on your existing Mongo apps.
Thanks everyone for your support so far! We believe this is the *first time* MongoDB apps will feel the speed of SingleStoreDB. You really get the scale of a distributed NoSQL database and the phenomenal analytics of an OLAP database. Icing on the cake is that SingleStoreDB can natively support vector functions that allow companies to build Generative AI applications!
Woww !!! SingleStore Kia for MongoDB looks promising. The Benchmark looks exciting! Can't wait to try it out and implement some use cases around it to bring event-based real-time data.
Thanks!
real-time even data is an excellent use case for this! Can't wait to see what you'll build.
This looks very helpful! Can’t wait to try it out. Congrats on the launch 🚀
Thanks much! Let us know if we can help you on the trial. Here's the link
Congratulations on the launch🎉
Appreciate that,
Wow, 100x faster analytics with SingleStore Kai sounds amazing! Just a quick question - how does the semantic search work in terms of complexity and accuracy? Great job on this!
Thanks. We will be posting a video from Jason on how to do this. Basically you can add another attribute in JSON that stores the embedding and then you can run DOT PRODUCT that returns the similarity match. Keep an eye on our website and our YouTube channel.
semantic search compares the 'meaning' of chunks of text to a search query, as the text was understood by an LLM. We'll have a video showing some of this up soon based on this example I tried it with book titles and descriptions (using OpenAI's text-embedding ada-002 model) and for example when I searched "funny astronaut stranded on mars has to survive, movie" I got the book The Martian as the top result, and when I searched "some guy rides along with a submarine captain classic french" I got 20,000 Leagues Under The Sea as the top result. Using it *feels* like using a search engine like Google or Bing, but in your own database! And thanks for the question and support!
Great job on the product launch! The product is fantastic and will help people boost their analytics by 100x!
Glad you think so...Thanks so much Ayesha!
It's great to see SingleStore Kai for MongoDB launch today! I'm excited to see how this lightning-fast API can power my apps and analytics!
Thanks for your support!
Cheers to the SingleStore team on the launch of Kai for MongoDB! The fusion of vector searches & high-speed JSON analytics looks game-changing. Exciting times ahead with faster and smarter apps!
Glad to see you're as excited as us about it!
123
Next
Last