Earlier today, Intel announced it has exceeded 500 AI models running on Intel CoreTM Ultra processors, which offer deep learning experiences, immersive graphics, and optimal battery life.
Intel’s investment in client AI, the AI PC transformation, framework optimizations, and AI tools including the OpenVINOTM toolkit led to this significant milestone. Among the 500 models, that can be deployed across CPU, GPU, and NPU, are OpenVINO Model Zoo, Hugging Face, ONNX Model Zoo, and PyTorch. Models include large language, diffusion, super-resolution, object detection, image classification/segmentation, and computer vision.
Why it’s important?
Intel’s achievement of 500 AI models is a milestone in its efforts to nurture and support the AI PC transformation. Aside from selling more units in one quarter than competing processors in 2023, Intel Core Ultra is the fastest-growing AI PC processor to date. Additionally, the platform provides the most AI models, frameworks, and runtimes of any processor vendor, making it the most robust for AI PC development.
AI-enhanced software features like object removal or image super-resolution rely on models. User-facing AI features can only be brought to market when there are enough enabled/optimized models. It is impossible to design a feature without a model. The feature cannot perform at its best without runtime optimization.
How it works?
In an AI-driven application, AI models play an important role in determining performance, stability, and capabilities. AI models are trained to analyze and infer large quantities of data, then act based on their conclusions. It is possible for developers to build new AI features for PCs using these models. There are more AI models, the more AI PC features there are.
Besides text, speech, audio, and images, AI models can also interact with other audiovisual sources found on PCs. For example, automatic text summarization, reducing energy consumption during teleconferencing, or removing unwanted objects from a picture are all AI-enhanced features. By load-balancing across all compute units, compressing the models for ease of running on an AI PC, and optimizing the runtime. OpenVINO takes advantage of Intel Core Ultra’s memory bandwidth and core architecture.
Providing an unmatched set of high-performance models on Intel Core Ultra opens up a wide field of innovation and simplifies the delivery of AI PC applications and features.
Over 500 AI models with optimized features and benefits:
-
Develops and deploys software more efficiently.
-
It encompasses more than 20 categories of artificial intelligence, including large language, diffusion, super-resolution, object detection, and computer vision.
-
This package includes Phi-2, Mistral, Llama, Bert, Whisper, and Stable Diffusion 1.5.
-
Enhances the stability, reliability, and performance of the system.
