Report an Error

AMD Radeon Instinct MI355X

Graphics Processor
CDNA Next
Cores
16384
TMUs
1024
ROPs
N/A
Memory Size
288 GB
Memory Type
HBM3e
Bus Width
8192 bit
GPU Chip
GPU
This product is not released yet.
Data on this page may change in the future.
The Radeon Instinct MI355X will be a professional graphics card by AMD, that is expected to launch in 2025. Built on the 3 nm process, and based on the CDNA Next graphics processor, the card does not support DirectX. Since Radeon Instinct MI355X does not support DirectX 11 or DirectX 12, it might not be able to run all the latest games. The CDNA Next graphics processor is a large chip with a die area of 1017 mm² and 185,000 million transistors. It features 16384 shading units, 1024 texture mapping units, and 0 ROPs. Also included are 1024 tensor cores which help improve the speed of machine learning applications. AMD has paired 288 GB HBM3e memory with the Radeon Instinct MI355X, which are connected using a 8192-bit memory interface. The GPU is operating at a frequency of 1000 MHz, which can be boosted up to 2400 MHz, memory is running at 2000 MHz.
Being a oam module card, the AMD Radeon Instinct MI355X does not require any additional power connector, its power draw is rated at 1400 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. Radeon Instinct MI355X is connected to the rest of the system using a PCI-Express 5.0 x16 interface.

Graphics Processor

GPU Name
CDNA Next
Architecture
CDNA 4.0
Foundry
TSMC
Process Size
3 nm
Transistors
185,000 million
Density
181.9M / mm²
Die Size
1017 mm²
Chip Package
MCM

Graphics Card

Release Date
2025
Announced
June 11th, 2025
Generation
Radeon Instinct
(MIx)
Predecessor
FirePro Data Center
Production
Unreleased
Bus Interface
PCIe 5.0 x16

Clock Speeds

Base Clock
1000 MHz
Boost Clock
2400 MHz
Memory Clock
2000 MHz
8 Gbps effective

Memory

Memory Size
288 GB
Memory Type
HBM3e
Memory Bus
8192 bit
Bandwidth
8.19 TB/s

Render Config

Shading Units
16384
TMUs
1024
ROPs
0
Compute Units
256
MCM
2
Tensor Cores
1024
L1 Cache
16 KB (per CU)
L2 Cache
16 MB
L3 Cache
256 MB

Theoretical Performance

Pixel Rate
0 MPixel/s
Texture Rate
2,457.6 GTexel/s
FP16 (half)
629.1 TFLOPS (8:1)
FP32 (float)
78.64 TFLOPS
FP64 (double)
78.64 TFLOPS (1:1)

Board Design

Slot Width
OAM Module
TDP
1400 W
Suggested PSU
1800 W
Outputs
No outputs
Power Connectors
None

Graphics Features

DirectX
N/A
OpenGL
N/A
OpenCL
3.0
Vulkan
N/A
Shader Model
N/A

Card Notes

All specs unknown

CDNA Next GPU Notes

Graphics/Compute: 9.4.3
Shader ISA: GFX9 (gfx940)
Video Core Next: 4.0.3
System DMA: 4.4.2
Jul 19th, 2025 12:30 CDT change timezone

New Forum Posts

Popular Reviews

TPU on YouTube

Controversial News Posts