Databricks expands reach in generative AI with MosaicML acquisition
In a significant industry development, Databricks, the renowned Data and AI company, has announced its definitive agreement to acquire MosaicML, a leading generative AI platform. This strategic move aims to make generative AI accessible to organizations of all sizes, empowering them to construct, possess, and safeguard generative AI models using their data. The acquisition, which includes retention packages, is valued at approximately $1.3 billion.
MosaicML has gained recognition for its cutting-edge MPT large language models (LLMs). With an impressive 3.3 million downloads of MPT-7B and the recent introduction of MPT-30B, MosaicML has demonstrated the ability of organizations to swiftly construct and train state-of-the-art models cost-effectively using their own data.
Prominent customers, including AI2 (Allen Institute for AI), Generally Intelligent, Hippocratic AI, Replit, and Scatter Labs, have leveraged MosaicML’s offerings for a diverse range of generative AI applications.
Databricks optimistic about the democratization of AI with MosaciML deal
Ali Ghodsi, Co-Founder and CEO of Databricks, emphasized the significance of enabling organizations to harness the potential of AI while retaining control over their data. Ghodsi expressed optimism regarding the democratization of AI through Databricks’ collaboration with MosaicML, uniting their shared vision rooted in transparency and a history of open-source contributions. The goal is to empower customers as they navigate the ongoing computing revolution.
“Every organization should be able to benefit from the AI revolution with more control over how their data is used. Databricks and MosaicML have an incredible opportunity to democratize AI and make the Lakehouse the best place to build generative AI and LLMs,” said Ali Ghodsi, Co-Founder and CEO of Databricks.
“Databricks and MosaicML’s shared vision, rooted in transparency and a history of open-source contributions, will deliver value to our customers as they navigate the biggest computing revolution of our time.”
Today, virtually every organization actively explores ways to maximize the benefits of generative AI and LLMs. Industry leaders are mainly concerned with leveraging these innovative technologies while controlling their most valuable asset: data. The ability to build, possess, and secure their models is paramount to organizations and executives alike.
Offering solutions to help customers control and secure their data without high costs
By combining MosaicML’s technology with the Databricks Lakehouse Platform, customers will access a user-friendly and rapid solution that ensures control, security, and ownership over their invaluable data without incurring high costs. MosaicML has reported that automatic optimization of model training can deliver training speeds 2x to 7x faster than standard methods.
Moreover, thanks to near-linear resource scaling, models with billions of parameters can be trained within hours rather than days. With Databricks and MosaicML, the cost of training and using LLMs will be reduced from millions to thousands of dollars.
The combined strength of Databricks’ unified Data and AI platform with MosaicML’s generative AI training capabilities will provide a robust foundation to serve the world’s largest organizations. Simultaneously, the platform’s flexibility will cater to a wide array of AI use cases, cementing Databricks’ position as a leader in the industry.
As the computing landscape evolves, Databricks and MosaicML are poised to spearhead the democratization of generative AI, enabling organizations to leverage these groundbreaking technologies while maintaining control over their data. The future of AI is becoming more accessible, empowering organizations worldwide to unlock the full potential of their data-driven initiatives.
