NVIDIA
GPUEmbeddedReleased —

NVIDIA RTX 5000 Ada Generation Embedded

16 GB GDDR6 · 256-bit · 9,728 cores

The RTX 5000 Ada Generation Embedded is an MXM 3.1 Type B module that brings NVIDIA's AD103-based Ada Lovelace workstation GPU to industrial, medical, robotics and aerospace systems, supplied through NVIDIA embedded partners with long-life availability. It carries 16 GB of ECC-capable GDDR6 and runs within a configurable 80-150 W total graphics power.

General

Manufacturer
NVIDIA
Model
RTX 5000 Ada Generation Embedded
Segment
Embedded
GPU chip
AD103
Architecture
Ada Lovelace
Release date
—

Memory

Memory size
16 GB
Memory type
GDDR6
Bus width
256-bit
Bandwidth
576 GB/s
Memory clock
2,250 MHz

Clocks & silicon

GPU clock
—
Boost clock
—
Shader units
9,728 CUDA Cores
Tensor / AI cores
304
RT cores
76
Transistors
45.9 billion
Technology (nm)
4 nm
TSMC 4N
TDP / TBP
150 W

Interface & APIs

Bus interface
PCIe 4.0 x16
SLI / multi-GPU
None
DirectX
12 Ultimate (12_2)
OpenGL
4.6
Vulkan
1.4
Other APIs
CUDA 8.9, OpenCL 3.0

AI support

Precisions
FP8FP16BF16INT8INT4TF32FP64
AI accelerator
4th-gen Tensor Cores (304)
Peak AI TOPS
—
Notes
Fourth-generation Tensor Cores add FP8 through the Transformer Engine alongside FP16, BF16, TF32, INT8 and INT4, with 2:4 structured sparsity. FP64 runs at one sixty-fourth of the FP32 rate.
All NVIDIA GPUs →

Embedded gpus from NVIDIA announced around the same time as the NVIDIA RTX 5000 Ada Generation Embedded. Add any of them to your compare list to see the specs side by side.