Reka selects Oracle Cloud Infrastructure to enhance AI model development

Reka, an AI research and product company known for its groundbreaking work in generative AI, has chosen Oracle Cloud Infrastructure (OCI) to support its continued innovation and expansion. This partnership aims to leverage OCI’s robust AI infrastructure, powered by NVIDIA GPUs, to advance the development and deployment of Reka’s multimodal AI models. These models, capable of understanding and generating content across more than 20 languages from inputs like text, code, images, and videos, will also be made available on the Oracle Cloud Marketplace.

Collaboration to Drive Multilingual Generative AI

Dani Yogatama, CEO and co-founder of Reka, emphasized the strategic importance of this collaboration: “We are partnering with Oracle to bring the power of generative AI to more organizations around the world,” said Yogatama. “OCI’s high-performance AI infrastructure capabilities, dedicated engineering support, and global footprint will continue to enable us to innovate and ship new models quickly and efficiently.”

Reka, founded by a team of experts from DeepMind, Google Brain, and FAIR, has developed advanced multimodal AI models that facilitate a wide array of applications. These include generating creative concepts, answering diverse questions, and deriving insights from large datasets. The collaboration with Oracle is set to enhance Reka’s ability to serve enterprises globally and cater to an expanding array of business use cases.

Leveraging Oracle’s Infrastructure for AI Innovation

Greg Pavlik, Senior Vice President of AI and Data Management Services at Oracle Cloud Infrastructure, also commented on the partnership’s potential. “Reka is pushing performance boundaries by training its multimodal foundation models on OCI Supercluster,” said Pavlik. “We are looking forward to working with the Reka team to deliver its multimodal models, with capabilities that stretch beyond text, to Oracle customers around the world.”

Reka’s models, including Reka Core and Reka Flash, will be accessible through private offers on the Oracle Cloud Marketplace. This platform serves as a centralized hub for enterprise applications developed by Oracle and its partners.

Oracle’s AI Infrastructure Paving the Way for Faster AI Deployments

The OCI’s purpose-built AI capabilities are already aiding many AI startups and enterprises like Reka in building and training models more efficiently. The OCI Supercluster can scale up to 4,096 OCI Compute bare metal instances equipped with 32,768 NVIDIA H100 Tensor Core GPUs. These are integrated via ultra-low-latency RDMA cluster networking, complemented by high-performance computing (HPC) storage options. Oracle’s commitment extends through the entire deployment process, with a dedicated engineering support team that assists customers from planning to launch, ensuring successful AI implementations.

The strategic partnership between Reka and Oracle represents a significant development in the AI industry, particularly in the area of generative AI models capable of processing multilingual and multimodal inputs. By utilizing Oracle’s powerful cloud infrastructure and NVIDIA’s advanced GPUs, Reka is set to enhance the speed, efficiency, and scalability of its AI model development. This move not only benefits Reka but also the broader ecosystem of users and developers engaged in AI-driven applications.