“OctoAI” provides a self-optimizing infrastructure that enables developers to run, tune and scale AI applications on any model
Seattle, June 14, 2023 /PRNewswire/ — OctoML Today we announced OctoAI, the industry’s first self-optimizing computing service for AI. The new platform provides developers with a fully managed cloud infrastructure designed to abstract the complexity of building and scaling AI applications. OctoAI allows you to run, tune and scale your model at will choose, Includes off-the-shelf open source software (OSS) and custom models. OctoAI gives developers easy access to cost-effective and scalable accelerated computing, allowing them to focus on building high-performance cloud-based AI applications and delivering a superior user experience to their customers. You will be able to
To help developers quickly build the latest and greatest models, OctoAI also introduces the world’s fastest and most affordable library of generative AI models leveraging the platform’s model acceleration capabilities. I’m here.OSS foundation model templates available at launch include Stable Diffusion 2.1, Dolly v2 and Llama 65BWhisper, Flan UL, Vicuña.
OctoML CEO Luis Ceze said, “AI is no longer a novelty, it’s a real business. But efficient computing is essential to make it happen.” “Companies are eager to build AI-powered solutions, but the process of moving models from development to production is highly complex and often requires expensive, specialized personnel and infrastructure. OctoAI is about making models work for the business, not the other way around.By abstracting away all the complexity, developers can build great applications without worrying about managing infrastructure. will be able to concentrate on
Ceze adds, “Our early OctoAI customers are using generative AI models like Stable Diffusion, FILM, and Flan UL to build a wide variety of applications. But they all have two things in common: First, customization is the cornerstone of delivering a unique experience to our customers.” Second, flexible hardware, from NVIDIA GPUs to specialized AI silicon like AWS Inferentia2. You need the ability to leverage your options and scale your services quickly. “
OctoAI features and benefits include:
- ease of use. Simplify deployment by choosing from a library of ready-to-use templates for popular open-source models. Select and customize (fine-tune) the model to meet your specific requirements. Easily integrate with your app and model development workflows.
- efficiency. Run, tune, and extend off-the-shelf open source software (OSS) and custom models. With automatic hardware selection, you Determine price/performance trade-offs.
- Freedom. Upgrade when new models come out. Bring your own custom model. No lock-in to models or services.
OctoML will host a virtual event today to unveil new services Wednesday, June 14th and 10am Pacific Time.To register, please visit this page.
About OctoML
OctoML is on a mission to make AI more accessible, sustainable, and harnessed to improve lives. Our platform, OctoAI, is a self-optimizing computing service that provides the most efficient AI infrastructure for open source and custom models. With OctoAI, developers get a managed infrastructure that rivals closed-model API vendors, deliver the best price/performance, and have the freedom to customize, integrate, and deploy future generative AI models.
Source OctoML
