Together AI: Advancing the Frontier of AI With Open Source Embeddings, Inference, and MongoDB Atlas
Blog post from MongoDB
Together AI, founded in San Francisco in 2022, aims to create a rapid cloud platform for generative AI (gen AI), supported by over $120 million in funding from investors such as Nvidia and Kleiner Perkins. The company focuses on open-source research and models, providing cloud services for developers and researchers to train and deploy gen AI models, emphasizing open and transparent AI systems. Recently, Together AI introduced the Together Embeddings endpoint to aid developers in building applications using retrieval-augmented generation (RAG), allowing gen AI models to produce more accurate, business-specific outputs. This service, integrating with MongoDB Atlas, LangChain, and LlamaIndex, offers access to leading open-source embedding models at a significantly lower cost than proprietary alternatives. A tutorial demonstrates how to build a RAG application using these integrations, showcasing a gen AI model that recommends Airbnb properties based on user criteria while maintaining factual accuracy. Meanwhile, MongoDB announced a leadership transition, as CEO Dev Ittycheria plans to retire and will be succeeded by Chirantan “CJ” Desai on November 10, 2025. Desai, known for his significant growth achievements at ServiceNow and Cloudflare, is expected to lead MongoDB into its next phase of evolution, MongoDB 3.0, as the company capitalizes on advancements in AI and data-intensive applications. Ittycheria will remain on the Board to ensure a smooth transition, expressing confidence in MongoDB's future under Desai's leadership and emphasizing the strategic timing of this change to foster continued growth and innovation.