Databricks and MosaicML Join Forces to Enable Organizations to Build and Secure Custom Generative AI Models

Introduction

Databricks, a prominent Data and AI company, has announced its definitive agreement to acquire MosaicML, a leading generative AI platform. The strategic partnership aims to make generative AI accessible to every organization, enabling them to build, own, and secure their own generative AI models using their proprietary data. With a transaction valued at approximately $1.3 billion, inclusive of retention packages, Databricks and MosaicML seek to democratize AI and establish the Databricks Lakehouse as the go-to platform for building generative AI models.

MosaicML's State-of-the-Art Generative AI Models

MosaicML is recognized for its cutting-edge MPT large language models (LLMs). The company's flagship models, including MPT-7B and the recently launched MPT-30B, have garnered significant traction, with over 3.3 million downloads. MosaicML's solutions enable organizations to rapidly develop and train state-of-the-art models using their own data in a cost-effective manner. Notable customers such as AI2, Generally Intelligent, Hippocratic AI, Replit, and Scatter Labs have leveraged MosaicML's platform for diverse generative AI use cases.

Enabling Organizations to Build, Own, and Secure Models

In today's AI landscape, every organization seeks to harness the potential of generative AI and large language models while maintaining control over their valuable data. Databricks and MosaicML understand this imperative and aim to provide organizations with a simple, fast, and cost-effective way to build, own, and secure their custom models. By leveraging MosaicML's technology, the Databricks Lakehouse Platform empowers customers to retain control, security, and ownership of their data without incurring exorbitant costs. MosaicML's automatic optimization of model training ensures significantly faster training times, ranging from 2x to 7x compared to standard approaches. With near-linear scaling of resources, organizations can train multi-billion-parameter models in hours rather than days, with costs amounting to thousands of dollars instead of millions.

Shared Vision and Integration of Expertise

The acquisition of MosaicML includes the integration of its entire team, including their renowned research team, into Databricks. MosaicML's research and engineering specialists are at the forefront of AI advancements, constantly enhancing model training efficiency. The team has contributed to popular open-source foundation models, such as MPT-30B, and has developed training algorithms that power MosaicML's products. Over time, MosaicML's platform will be seamlessly integrated into Databricks' unified Data and AI platform, providing customers with a comprehensive and unified solution to build, own, and secure their generative AI models. The acquisition empowers organizations to exercise greater choice, train models using their unique data, and create differentiated intellectual property to drive business success.

Advancing the AI Journey Together

MosaicML's mission aligns with Databricks' vision of democratizing AI and creating a transparent and open-source ecosystem. By joining forces, both companies aim to drive value for customers as they navigate the transformative AI revolution. The acquisition marks a significant step towards making generative AI accessible to a wide range of organizations, enabling them to leverage their data effectively and drive innovation in their respective industries. Databricks and MosaicML are committed to collaborating with the AI community and are excited about the opportunities ahead.

Conclusion

Databricks' acquisition of MosaicML is set to revolutionize the generative AI landscape by democratizing access to cutting-edge models and ensuring organizations have control over their data. By combining the strengths of Databricks' Lakehouse Platform with MosaicML's expertise in generative AI training, the partnership will deliver a unified platform capable of serving both large organizations and addressing a broad range of AI use cases. The acquisition reflects Databricks' and MosaicML's shared commitment to transparency, data control, and empowering organizations to build their own generative AI models. As the transaction proceeds, Databricks looks forward to integrating MosaicML's capabilities and continuing the journey with the AI community, driving innovation, and solving the world's most challenging problems.

Key Takeaway

Databricks' acquisition of MosaicML signals a significant step towards democratizing generative AI and enabling organizations to build and secure their own custom models. Through the integration of MosaicML's advanced training capabilities into the Databricks Lakehouse Platform, customers gain unprecedented control, security, and ownership over their data-driven AI models. The partnership solidifies Databricks' position as a leader in the AI industry, driving innovation, and empowering organizations to harness the power of generative AI.