AI innovators around the world use Oracle Cloud Infrastructure (OCI) AI infrastructure and OCI superclusters to train AI models and deploy AI inference and applications. Fireworks AI, Hedra, Numenta, Soniox, and hundreds of other AI innovators have chosen OCI to control their scalability, performance, cost-effectiveness, selection of computing instances, and where AI workloads run.
As the industry is rapidly embracing AI to promote innovation and efficiency, the AI companies offering these services need reliable, secure, highly available cloud and AI infrastructure, allowing GPU instances to be scaled quickly and economically. The OCI AI infrastructure allows AI companies to access high-performance GPU clusters and the scalable computing power needed for AI training, AI inference, digital twins, and highly parallel HPC applications.
“Among AI innovators, OCI has quickly become the perfect destination for training and speculation needs of all sizes,” said Chris Gandolfo, executive vice president of Oracle Cloud Infrastructure and AI. “OCI AI infrastructure provides the ultra-fast networking, optimized storage, and cutting-edge GPUs that AI companies rely on to withstand the next wave of innovation.”
Global AI Innovators choose Oracle
Fireworks AI is an inference platform that enables developers and businesses to build highly optimized, production-ready, generic AI applications, offering over 100 cutting-edge open models in text, images, audio, embedding and multimodal formats. Fireworks AI will help expand its services globally by providing over 2 trillion inference tokens on the platform daily, using bare metal instances accelerated by OCI computing with Nvidia Hopper GPUs and AMD MI300X GPUs.
“We're committed to providing a range of services to our customers,” said Lin Qiao, co-founder and CEO of Fireworks AI. “OCI AI infrastructure allows developers to provide the ultra-fast response times and production-grade stability they expect. It allows AI workloads to be efficiently handled, minimized downtime, and ensures AI applications can run smoothly at scale.
Hedra, an AI-driven video creation company, allows users to create videos with characters like life. By deploying multimodal foundation models for image, video and audio of OCI Compute bare Metal instances accelerated by the Nvidia Hopper GPU, Hedra has reduced GPU costs, increased training speeds, and reduced model iteration times.
“We're committed to providing a range of services to our customers,” said Michael Lingelbach, founder and CEO of Hedra. “OCI handles model training and guesswork across video, audio and image data, providing the rapid processing needed to render real-time character and meet the high storage needs of large datasets. This allowed us to release the latest models, character 3, and content creation platform, Hedra Studio, quickly and without hitch.”
Numenta is an AI technology company focused on maximizing the performance and efficiency of deep learning systems. By using OCI computing bare metal instances accelerated by Nvidia GPUs, Numenta has access to a variety of reliable, high-performance training instances, achieving faster training speeds and increased learning cycles.
“OCI provides the high-performance infrastructure and powerful operational support needed to push the boundaries of AI without compromising speed or accuracy,” said Dan Steele, CEO of Numenta. “OCI allows us to confidently accelerate the development of next-generation technology and take a major step with efficient IntelligencetM.”
Soniox is an AI company at the forefront of audio and speech AI, and is a Pioneers Foundational AI model for audio, speech and language understanding. With a new universal multilingual voice AI model hosted by OCI, Soniox uses the OCI Compute Bare Metal Instance accelerated by the Nvidia Hopper GPU to train its model to seamlessly recognize and understand 60 languages in 60 languages with low latency and higher accuracy.
“We're committed to providing a range of services to our customers,” said Klemen Simonic, Founder and CEO of Soniox. “OCI has access to the latest AI innovations that can push the boundaries of voice recognition and audio understanding while significantly reducing deployment time and operational costs.”
