Intel Gaudi 3 Finally Releasing Next Week

Intel’s plans for 2024 are going strong. The company announced that Intel Gaudi 3 accelerator, its AI hero capable of scaling without limits built to handle demanding training and inference, is set to release next week. 

Pat Gelsinger, the CEO of Intel had teased the accelerator last year at its Intel AI Everywhere event last year.

Introducing #IntelGaudi 3 accelerator – an AI hero capable of scaling without limits. Coming to #GenAI workloads everywhere on 09.24.24. #StartswithIntel https://t.co/D3hE75KwN0

— Intel (@intel) September 19, 2024

Later, Intel had again announced Gaudi 3 at the Intel Vision 2024 event, in a bid to keep pace with the growing demand for semiconductors capable of training and deploying large AI models. The company claims that it boasts over double the power efficiency compared to its predecessor and is capable of running AI models 1.5 times faster than NVIDIA’s H100 GPU.

In comparison to NVIDIA, Intel claims its chips consume less power. NVIDIA currently dominates approximately 80% of the AI chip market with its GPUs, which have been the preferred choice for AI developers in the past year.

Meanwhile, NVIDIA is already a generation ahead of Intel with its Blackwell release. However, it has been delayed caused by a mask issue affecting chip yield, not design flaws. On the other hand, Blackwell promises substantial AI performance improvements over competitors like AMD’s MI300X, which the company is also planning to push to the market soon. 

Given the time and date of release of the products, Intel now has the upperhand to pose a strong competition to both the NVIDIA’s Blackwell and AMD’s MI300X.

Intel Gaudi 3 AI accelerators are designed to supercharge AI applications, supporting state of the art generative AI and LLMs in data centres. Paired with Intel Xeon processors, Gaudi 3 delivers enterprise grade performance and reliability across various use cases, including 3D generation, language translation, sentiment analysis, and text to image generation. 

In 2025, the successor to Gaudi3, Falcon Shores, will merge the AI capabilities of Gaudi with the powerful GPUs from Intel, all within a single package. This is something that would give Intel the edge over others.

The AI accelerator was tested on models like Meta’s open-source Llama and the Falcon by TII. Gaudi 3 demonstrated its capability to facilitate the training or deployment of models such as Stable Diffusion or OpenAI’s Whisper model for speech recognition.