Marqo

We had the pleasure of attending the Sydney AI Meetup to present a collaboration between Marqo and Amazon Web Services (AWS)! Together, we demonstrated an advanced video search and Q&A solution that leverages Marqo's embedding model and vector database alongside Amazon Bedrock LLM. This solution enables users to effortlessly find precise video content and timestamps, transforming productivity by removing the hassle of manually searching through entire videos for key information. A huge shoutout to Marqo's Solutions Architect, Owen Elliott, for presenting as the guest speaker at the event. Find out more: https://lnkd.in/eQwD7kEK

We're one week away from #AWSreinvent, one of the largest and most anticipated cloud computing events of the year! Join us at booth 1897?to experience firsthand how Marqo's cutting-edge AI-powered search and recommendation solutions can transform your business. Whether you’re looking to enhance your product search capabilities, improve user experiences, or leverage semantic search for your RAG application, Marqo.ai is here to help you build. Stop by booth 1897?for a personalized demo and conversation, or feel free to schedule a meeting in advance: https://lnkd.in/ebbTCBh7

Want to visit #TheSphere at AWS #Reinvent this year? Join?us?for dinner, drinks, and "Postcard From Earth," a breathtaking visual journey directed by Darren Aronofsky (Requiem for a Dream, Black Swan, Pi), and an immersive experience combines stunning visuals and captivating storytelling shown on an 18K-resolution display. Space is limited, so be sure to register early! Register here --> https://lnkd.in/gpN9vDev

We're on Bluesky! ?? Personalised feeds, starter follower packs, and lists? This platform is next level for creative content sharing. Check out our Bluesky Social profile to get started and follow the industry's leading ML & AI professionals and posts. Come say hi and follow us here: https://lnkd.in/d4zHyZrD Or drop your handle in the comments and we'll give you a follow.

Building a search application has never been so easy ?? Marqo has got you covered from training, to inference, to storage. You don't need to calculate the vectors yourself, simply select the model you want to use. Marqo supports hundreds of embedding models out of the box, as well as custom weights, and models fine-tuned with Marqtune (Marqo's Generalized Contrastive Learning framework). Find out more: https://lnkd.in/dKRiyzqs

5 new YouTube videos on ML & AI just dropped ???? One thing I love about working at Marqo is being able to teach people how to build awesome projects and learn more about ML & AI. These videos focus on how you can make the most out of Marqo's new open source, state-of-the-art embedding models. These models are open source on Hugging Face ?? Videos include... ?? How to build your own image search (multimodal) application... - Using Marqo for unified embedding generation and vector search engine - Using Gradio for the UI - 2 videos, one on running in Google Colab and the other running from GitHub ?? A breakdown of Marqo's state-of-the-art embedding models - What are embedding models? - Why did Marqo create them specifically for ecommerce? - How much better do these state-of-the-art models perform? - Where can I find these models and use them? ?? How to build an Image Classification application - Video 1: Building an image classification app using Hugging Face transformers - Video 2: Building an image classification app using the open-clip library in Python Full playlist here: https://lnkd.in/ekdYgJKi

We just launched 5 new Marqo YouTube videos ?? + Analysis and breakdown of Marqo's state-of-the-art ecommerce embedding models: https://lnkd.in/eRU9TeES + How to perform image classification with Hugging Face transformers and Marqo's embedding models: https://lnkd.in/eG7Z8Gag + How to perform image classification with OpenAI's CLIP library and Marqo's embedding models: https://lnkd.in/euYDvDuv + How to build a search application with Marqo and Gradio (both in Google Colab and GitHub): GitHub: https://lnkd.in/ecq6uWFv Google Colab: https://lnkd.in/ekbTs54p Visit our YouTube to check them out: https://lnkd.in/eUTu4Bkg

Ecommerce Search and Classification Just Got Better with Marqo... In this demo, we put two models (Marqo vs OpenAI) to the test for taxonomy classification—an essential task for organizing products in ecommerce. Marqo's SOTA ecommerce embedding model delivers precise classifications that align with real-world product taxonomies. OpenAI's ViT-B-16 model, while general-purpose, struggles to keep up with the nuanced demands of ecommerce. Why does this matter? + Accurate taxonomy classification enhances search relevance. + Organized categories improve the shopping experience. + Tailored models deliver results that generic ones just can't match. At Marqo, we provide a platform for you to train and deploy embedding models specific to your use-case. Providing the tooling you need to accelerate development, reduce risk and improve results. If you want to improve your search, book a demo and get $500 free Marqo Cloud credits: https://lnkd.in/eeVUx_Jk

What a week for Marqo on Hugging Face! 1. We released two of the best embedding models for #Ecommerce Search and Recommendations available anywhere: Marqo-Ecommerce-L & Marqo-Ecommerce-B. 2. We hit 100k downloads in one month on our state-of-the-art multimodal embedding model, fine-tuned for #fashion, marqo-FashionSigLIP. 3. Our evaluation datasets released with our ecommerce embedding model launch have been trending on the Hugging Face dataset hub. For updates on new model and dataset releases, follow us on Hugging Face: https://lnkd.in/e3RU44Gn Read more about the ecommerce embedding models: https://lnkd.in/eiunJ4qT

Creating an Image Search Application for Ecommerce can be a challenging task, but?Marqo Cloud and our new, state-of-the-art ecommerce embedding models make it easy. In this blog, we walk you through the entire process... 1. Setting up your Marqo index. Marqo handles the embedding generation so all you need to do is specify which models you want to use. In this case, we load them directly from Hugging Face. 2. Adding documents to your Marqo index. We use a subset from the Marqo-GS-10M dataset. One of the largest and richest datasets for multimodal product retrieval. Available on Hugging Face. 3. Building an easy-to-set-up user interface using Gradio. We build an image search UI where users can input a query as well as what they want to see 'more' or 'less' of. 4. Optionally deploy to Hugging Face Spaces with Gradio. This article gives you the option to deploy directly to Hugging Face Spaces, giving you a live working demo that you can share with your colleagues and friends. 5. Managing your index once set up. Finally, we give you tips on managing and cleaning your index. Blog post: https://lnkd.in/eWqH_7gv GitHub: https://lnkd.in/ekVRDBuV Marqo on Hugging Face: https://lnkd.in/e3RU44Gn