The new AMD Ryzen™ Threadripper™ 9000 Series processors for high-end desktops (HEDT), previously showcased at Computex, will start shipping on 31 July 2025.
Based on the new “Zen 5” CPU architecture, the Threadripper 9000 Series will be available from AMD’s wide network of regional system builders and channel partners next month, including the new Threadripper PRO 9000 WX-Series processors.
Threadripper 9000 Series HEDT processors are aimed at creators, AI developers and enthusiasts, who are looking to accelerate their rendering, 3D content creation, video editing and AI processing with unparalleled multi-core performance, cutting-edge memory capabilities and expansive I/O connectivity that push beyond traditional desktop limits.
Threadripper 9000 Series processors deliver overclockable desktop power with up to 64 “Zen 5” CPU cores and 128 threads, up to 256MB of L3 cache and 4-channel DDR5-6400 memory*. Compared to its predecessor, AMD have enhanced the support for AVX-512 with a full 512-bit data path and now offer up to 80 PCIe® 5.0 lanes to ensure users get the most out of the latest high-end GPUs.
Building the Ultimate High-End Desktop
Content creators and video editors find every task – whether it’s rendering, compiling, AI inference or multitasking – demands a system that handles heavy workloads without compromise, and investing in the right platform is key to maximising efficiency.
The Threadripper 9000 Series for high-end desktops is available with 24, 32 or 64 cores, and now offers up to 80 PCIe® 5.0 lanes for graphics, storage, and more (up from 48 lanes before). Capable of twice the memory bandwidth of typical dual-channel desktop systems, the quad-channel DDR5 memory controller on Ryzen Threadripper 9000 Series processors can support even the most intensive workflows.
|
Processor |
Cores / Threads |
Boost (maximum frequency achieved by a single core) /Base |
Total |
TDP |
|
AMD Ryzen Threadripper 9980X |
64 / 128 |
Up to 5.4 / 3.2 GHz |
256MB |
350W |
|
AMD Ryzen Threadripper 9970X |
32 / 64 |
Up to 5.4 / 4.0 GHz |
128MB |
350W |
|
AMD Ryzen Threadripper 9960X |
24 / 48 |
Up to 5.4 / 4.2 GHz |
128MB |
350W |
In AMD internal testing, the AMD Ryzen™ Threadripper™ 9000 Series processors have been shown to outperform Intel® Xeon® W-3500 series processors across many different workloads, enabling AI developers and software engineers to benefit from shorter compilation times and faster AI processing, while creators, product designers and CAD engineers can execute heavy rendering jobs while multi-tasking with complex 3D models.
AMD Ryzen™ Threadripper™ 9000 Series high-end desktop processors are ideal for local AI processing, rendering, heavy 3D content creation, VFX, and software development thanks to support for up to 80 PCIe 5.0 lanes which provide maximum performance for advanced set-ups using multiple GPUs and NVMe storage.
Threadripper Motherboard Platform Feature Support
AMD Ryzen Threadripper 9000 Series processors use the same sTR5 socket as the previous generation and maintain a TDP of 350W. This makes them compatible with existing Threadripper motherboards which feature support tailored to meet the distinct needs of enthusiasts and professionals alike.
|
Chipset |
Processors |
Memory Support |
PCIe® lanes |
Features |
|
AMD WRX90 |
Ryzen Threadripper PRO 9000 WX-Series |
8-Channel |
148 / 144 |
AMD PRO Technologies Overclocking supported (not on OEM systems)* |
|
AMD TRX50 |
Ryzen Threadripper 9000 Series & Ryzen Threadripper PRO 9000 WX-Series |
4-Channel |
92 / 88 |
Overclocking supported* |
High-End PCIe® 5.0 GPU Support for AI and Machine Learning
Threadripper 9000 processors are ideal for local AI fine-tuning, inference and application development thanks to support for up to 80 PCIe 5.0 lanes which provide the required bandwidth for advanced set-ups using one or more high-end GPUs.
Built to accelerate medium to large AI models directly on local desktops, the new AMD Radeon™ AI PRO R9700 is an ideal match for advanced Threadripper 9000 desktop systems. It combines 32GB of high-speed GDDR6 memory, is based on the new RDNA™ 4 architecture, and offers broad support for modern AI data types. 64 Compute Units, 128 second-generation AI accelerators, and FP8, FP16, and INT8 precision equip developers with the compute power needed to handle today’s most demanding AI workloads.
* Overclocking AMD processors’ warranties will be voided if they are used outside of their prescribed use and specifications, for example altering clock frequencies and/or multipliers or altering memory timing and/or voltage. This may also void warranties offered by the system manufacturer or retailer. Users assume all risks and liabilities that may arise out of overclocking AMD processors, including, without limitation, failure of or damage to hardware, reduced system performance and/or data loss, corruption or vulnerability. GD-106
