AMD Unveils Radeon AI PRO R9700 Graphics Card at Computex 2025

AMD has announced the Radeon AI PRO R9700, a powerful graphics card designed to boost professional visualization and Artificial Intelligence workloads with up to 128 GB of scalable memory.

At Computex 2025, AMD introduced the Radeon AI PRO R9700 graphics card, targeting professionals working in visualization and Artificial Intelligence acceleration at the edge. Built on the advanced 4 nm ´Navi 48´ silicon, the R9700 offers 64 compute units and 128 dedicated AI accelerators, establishing itself as a high-performance variant of the Radeon RX 9070 XT desktop card. The graphics card is equipped with 32 GB of fast 20 Gbps GDDR6 memory paired with a 256-bit interface, enabling a robust 640 GB/s memory bandwidth. With a total board power rating of 300 W, AMD claims the card can achieve computational throughput of up to 96 TFLOPs for FP16 operations and 1,531 AI TOPS (INT4 sparse).

One of the standout features of the Radeon AI PRO R9700 is its focus on memory capacity and scalability, aimed squarely at handling large-scale Artificial Intelligence models. AMD highlights that its 32 GB memory—double that of the NVIDIA GeForce RTX 5080—makes it especially suitable for demanding workloads, and may give clues about competitive pricing strategies. The R9700 supports driver-level scalability for up to four GPUs via PCIe Gen 5, creating a pool of up to 128 GB of graphics memory. This expanded memory pool is particularly advantageous for organizations deploying intricate models like Mistral Large Instruct with 123 billion parameters or DeepSeek R1 Distill, which is based on Llama´s 70 billion parameters. Both models typically require memory footprints ranging from 112 GB to 116 GB, making the R9700 well-suited for researchers and engineers working at the forefront of machine learning and inference.

Additionally, AMD announced that the Radeon AI PRO R9700 would launch alongside its new Ryzen Threadripper PRO 9000 ´Shimada Peak´ workstation processors, with product availability slated for July 2025. By providing extensive GPU memory, significant AI acceleration capabilities, and support for large model inference, AMD aims to empower organizations navigating the rapidly growing Artificial Intelligence sector and professional visualization landscapes. The R9700 is positioned as a compelling solution for next-generation edge workloads and high-end compute tasks, emphasizing the company’s commitment to serving the needs of modern professionals.

76

Impact Score

Artificial Intelligence LLM confessions and geothermal hot spots

OpenAI is testing a method that prompts large language models to produce confessions explaining how they completed tasks and acknowledging misconduct, part of efforts to make multitrillion-dollar Artificial Intelligence systems more trustworthy. Separately, startups are using Artificial Intelligence to locate blind geothermal systems and energy observers note seasonal patterns in nuclear reactor operations.

Saudi Artificial Intelligence startup launches Arabic LLM

Misraj Artificial Intelligence unveiled Kawn, an Arabic large language model, at AWS re:Invent and launched Workforces, a platform for creating and managing Artificial Intelligence agents for enterprises and public institutions.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.