Test out our new, state-of-the-art fashion embedding models directly on Hugging Face Spaces 🤗 Here's a short video of our Hugging Face Space in action! Want to give it a go yourself? Try it out here👇 https://lnkd.in/eXb2cqKf
Marqo
Technology, Information and Internet
San Francisco, California 6,949 followers
Unified embedding generation and search engine.
About us
Marqo is more than a vector database, it's an end-to-end vector search engine for both text and images. Vector generation, storage and retrieval are handled out of the box through a single API.
- Website
-
https://www.marqo.ai/
External link for Marqo
- Industry
- Technology, Information and Internet
- Company size
- 11-50 employees
- Headquarters
- San Francisco, California
- Type
- Privately Held
Locations
-
Primary
515 Mission St
San Francisco, California 94105, US
-
276 Flinders St
Melbourne, Victoria 3000, AU
-
3 Waterhouse Square
London, EC1N 2SW, GB
Employees at Marqo
Updates
-
Transforming Ecommerce Search with Marqo’s Hybrid Search! 🔎🚀 We're excited to share our latest article where we dive into the power of Hybrid Search using Marqo. In this post, we walk through building a multimodal search system that leverages both vector and lexical search, providing a seamless way to enhance search accuracy and relevance in ecommerce 👔👗 From implementing tensor and lexical retrieval methods to experimenting with Reciprocal Rank Fusion (RRF) and sponsored products, we cover all the essentials to advance your search capabilities. This article contains practical examples, detailed code snippets, and a hands-on UI to see the impact of different search configurations in real-time 🕰️🔥 Whether you’re working with complex queries or just looking to improve your site's search experience, give the article a read or chat with one of our team. Article: https://lnkd.in/evta8_EJ Book a Demo: https://lnkd.in/eWTjHWdx
-
-
Last week we released our new Ecommerce & Retail embedding models: Marqo-FashionCLIP and Marqo-FashionSigLIP 👔👗 In our latest article, we dive into a simple Ecommerce Image Classification Tutorial with Marqo-FashionCLIP 🔥 Read the article here: https://lnkd.in/d6y22MdQ Benchmarks: https://lnkd.in/evPFQAwZ Hugging Face: https://lnkd.in/d5Hui-aw GitHub: https://lnkd.in/e7AiaB79
-
-
Try out our new, state-of-the-art, multimodal embedding models for Ecommerce & Retail directly from Hugging Face 🤗🔥: https://lnkd.in/e5TMyrsx For more information on the models, visit our article: https://lnkd.in/gZbJTmYv Hugging Face: https://lnkd.in/eK8Gtg2z GitHub: https://lnkd.in/ePe8GUR2 Join our community: https://lnkd.in/e2hxAnJR
-
-
Sign up to our newsletter to receive the latest announcements, events and updates from Marqo! Newsletter Sign Up: https://buff.ly/4chl3Gs You can also join our Community: https://buff.ly/4cIbKjZ
-
-
Marqo reposted this
New open source state-of-the-art embedding models for fashion - Marqo-FashionCLIP and Marqo-FashionSigLIP: - Outperform FashionCLIP2.0, and OpenFashionCLIP on all benchmarks (up to +57%). - Evaluated across 7 publicly available fashion datasets and three task types. - 10% faster inference. - Optimized over multiple fashion specific attributes including colors, style, materials, details, categories and keywords. - Open source - Apache 2.0 license. - Download from Hugging Face or use directly in Marqo. Marqo-FashionCLIP & Marqo-FashionSigLIP are two new state-of-the-art multimodal embedding models for search and recommendations in the fashion domain. Marqo-FashionCLIP & Marqo-FashionSigLIP can produce embeddings for both text and images that can then be used in applications like search and recommendations. The models were trained on millions of up-to-date fashion products with rich meta-data containing detailed descriptions, colors, styles, keywords and materials. The loss consisted of seven parts to optimize specifically for long descriptions, product titles, colors, materials, categories, details and keywords. This results in a model that is able to retrieve more relevant results for both keyword-like short text and longer descriptive text! Learn more: Blog post: https://lnkd.in/gwjt5nMp Hugging Face: https://lnkd.in/grRU86Wg Marqo: https://lnkd.in/gsFRwZT5 GitHub for benchmarks: https://lnkd.in/gAa-99d7 Feel free to let me know any feedback or other embedding models you are interested in! Great work from David Jung!
-
-
Marqo reposted this
Marqo have just announced a new family of open-source, multimodal embedding models for Ecommerce and Retail 🤗 🔥 You can run them on Google Colab here: https://lnkd.in/e9KGQ8Pi They are 150M parameter embedding models that: 🚀 Outperform FashionCLIP2.0, and OpenFashionCLIP on all benchmarks (up to +57%). 🚀 Are evaluated across 7 publicly available fashion datasets and three task types. 🚀 Are optimized over multiple fashion specific attributes including colors, style, materials, details, categories and keywords. 🚀 Can be downloaded from Hugging Face or used directly, with fully managed inference and vector retrieval, on Marqo Cloud. Hugging Face: https://lnkd.in/eJRmDs9k GitHub: https://lnkd.in/e75hP5M8 Read more about these models in this article: https://lnkd.in/e4Cwpidj #huggingface #machinelearning #ML #AI #embedding #ecommerce
-
-
We are thrilled to be releasing 2 open-source, state-of-the-art, multimodal embedding models for search and recommendations in the Ecommerce and Retail domain 🔥🚀 Marqo-FashionCLIP & Marqo-FashionSigLIP are 150M parameter embedding models that: 🌟 Outperform FashionCLIP2.0, and OpenFashionCLIP on all benchmarks (up to +57%). 🌟 Outperform ViT-B-16-SigLIP in all benchmarks (Marqo-FashionSigLIP only). 🌟 Are 10% faster for inference than FashionCLIP2.0, and OpenFashionCLIP. 🌟 Use Generalized Constrastive Learning (GCL) to optimize over seven fashion specific aspects including descriptions, titles, colors, details, categories, keywords and materials. Download our models directly from Hugging Face and use them anywhere with our reference implementation. You can also deploy it in Marqo Cloud. Find out how in our article! Full article: https://lnkd.in/evPFQAwZ Hugging Face: https://lnkd.in/eK8Gtg2z GitHub: https://lnkd.in/e7AiaB79
-
📣 Join the Marqo Community! We've had incredible conversations with our community, from innovative Marqo project ideas to insightful feature requests. Our community is a vibrant hub for growth and learning. If you haven't already, come and say hello! Slack Community: https://buff.ly/4cIbKjZ
-