Nvidia AI chip AMD launches GPU 'accelerator' MI300
Economic Observer Network reporter Qian Yujuan
When NVIDIA broke into the "Trillion Dollar Club" with its computing power dividend, its old rival AMD (American Advanced Micro Devices) also launched its latest GPU product.
On June 13th, Eastern Time, AMD CEO Su Zifeng introduced this AI chip called an "accelerator" - MI300. Its memory far exceeds the 120GB NVIDIA GPU chip H100, up to 192GB. , In addition, a single MI300 can also support model operation with 80 billion parameters. In terms of achieving ultra-high model loading on a single GPU, AMD is undoubtedly "harder" than Nvidia.
By increasing the memory and computing power of AI chips, AMD shows that its innovative "accelerator" can ensure the processing speed of chatbots such as ChatGPT and large language models using generative AI. Different from NVIDIA CEO Jen-Hsun Huang's demonstration of AI composition at Computex Taipei, in order to show the capabilities of his own AI chips, Su Zifeng asked a large language model to create a poem about San Francisco at the conference in San Francisco.
Su Zifeng is optimistic about the opportunities in the future data center market. She predicts that the entire data center and AI accelerator market will reach about US$30 billion this year, but by 2027 the market size may exceed US$150 billion, with an annual growth rate exceeding US$150 billion. 50%.
Although AMD's first quarter financial report was the first revenue decline in four years, its share of the data center market reached 25.8%, which is also a record high since the development of AMD's data center business.
According to Su Zifeng’s summary, AMD further accelerated its artificial intelligence roadmap this quarter and implemented priority strategies for data center and embedded businesses.
Regarding the price of MI300, Su Zifeng did not announce it. She previously responded to investors at AMD’s earnings call that this AI product will begin to generate sales revenue in the fourth quarter of this year and will not be until 2024. "more meaningful".
In addition to releasing good news for the future, when AMD announced on its official website that it will showcase the latest data center and artificial intelligence products at an event called "AMD Data Center and Artificial Intelligence Technology Premiere", its stock price rose. There was a volatile increase, with an increase of nearly 10%.
Although the capital market had high hopes for AMD's release in the early stage, AMD's intraday stock price fell significantly during the official event on June 13, Eastern Time, closing at $124.53 per share, a decrease of 3.61%; on the other hand, its opponents The stock price of Nvidia (code: NVDA) fluctuated and rose during the same period, closing at $410.22 per share, an increase of 3.90%.
Before its official launch, MI300 was regarded by AMD as the "ultimate weapon" to compete with Nvidia's AI chips. Su Zifeng even put pressure on this "accelerator" product, "which can help the company occupy the market." , meeting all the requirements of the AI and supercomputing ecosystem.”
The above is the detailed content of Nvidia AI chip AMD launches GPU 'accelerator' MI300. For more information, please follow other related articles on the PHP Chinese website!

Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

Video Face Swap
Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Article

Hot Tools

Notepad++7.3.1
Easy-to-use and free code editor

SublimeText3 Chinese version
Chinese version, very easy to use

Zend Studio 13.0.1
Powerful PHP integrated development environment

Dreamweaver CS6
Visual web development tools

SublimeText3 Mac version
God-level code editing software (SublimeText3)

Hot Topics











Hey there, Coding ninja! What coding-related tasks do you have planned for the day? Before you dive further into this blog, I want you to think about all your coding-related woes—better list those down. Done? – Let’

Introduction OpenAI has released its new model based on the much-anticipated “strawberry” architecture. This innovative model, known as o1, enhances reasoning capabilities, allowing it to think through problems mor

Introduction Imagine walking through an art gallery, surrounded by vivid paintings and sculptures. Now, what if you could ask each piece a question and get a meaningful answer? You might ask, “What story are you telling?

Introduction Mistral has released its very first multimodal model, namely the Pixtral-12B-2409. This model is built upon Mistral’s 12 Billion parameter, Nemo 12B. What sets this model apart? It can now take both images and tex

SQL's ALTER TABLE Statement: Dynamically Adding Columns to Your Database In data management, SQL's adaptability is crucial. Need to adjust your database structure on the fly? The ALTER TABLE statement is your solution. This guide details adding colu

Troubled Benchmarks: A Llama Case Study In early April 2025, Meta unveiled its Llama 4 suite of models, boasting impressive performance metrics that positioned them favorably against competitors like GPT-4o and Claude 3.5 Sonnet. Central to the launc

Can a video game ease anxiety, build focus, or support a child with ADHD? As healthcare challenges surge globally — especially among youth — innovators are turning to an unlikely tool: video games. Now one of the world’s largest entertainment indus

While working on Agentic AI, developers often find themselves navigating the trade-offs between speed, flexibility, and resource efficiency. I have been exploring the Agentic AI framework and came across Agno (earlier it was Phi-
