GeekRadar / The comparison desk
COMPARE.
GPU
NVIDIA

NVIDIA Jetson AGX Orin 32 GB

Graphics processor

Video memory
32 GB*
Memory bus
256 bit*
Rated memory bandwidth
204.80 GB/s*

No matched benchmark results are available for this entry yet.

Processing & platform

Architecture
Ampere*
Process node
8 nm*
GPU cores
1,792*
Graphics interface
PCIe 4.0 x4*
Reported chip die area
200 mm²*
GPU L2 cache
6 MB*
Graphics processor
GA10B*
Texture mapping units
56*
Raster output units
24*
Streaming multiprocessors
14*
Tensor cores
56*
Chip foundry
Samsung*
Card or module format
IGP*
Declared GPU TDP
40 W*
DirectX / feature level
12 Ultimate (12_2)*
OpenGL
4.6*
OpenCL
3.0*
Vulkan
1.4*
CUDA compute capability
8.7*
Shader model
6.8*

Memory & storage

Memory options
32*
Video memory
32 GB*
Video memory type
LPDDR5*
Memory bus
256 bit*
Rated memory bandwidth
204.80 GB/s*

Power specifications

Thermal design power
40 W*

System configuration

Detailed specifications

Specifications 1 · 24 fields
proc die size
200* mm²
cores
1,792*
tensor cores
56*
sm count
14*
memory clock
800*
ram
32*
name
NVIDIA Jetson AGX Orin 32 GB*
ram type
LPDDR5*
bus width
256*
bus interface
PCIe 4.0 x4*
ram bandwidth
204.8* GB/s
fp16 performance
6.666*
fp32 performance
3.333*
tdp
40* W
slot width
IGP*
api cuda
8.7*
is retail board
No*
gpi value
4*
vendor
nvd*
manufacturer
NVIDIA*
arch name
Ampere*
card release date
2023-02-11*
proc foundry
Samsung*
proc process size
8* nm
Specifications 2 · 35 fields
CUDA compute capability
8.7*
DirectX / feature level
12 Ultimate (12_2)*
OpenCL
3.0*
OpenGL
4.6*
Shader model
6.8*
Vulkan
1.4*
Architecture
Ampere*
Graphics interface
PCIe 4.0 x4*
Memory bus
256* bit
Reported shader processors
1,792*
Reported theoretical FP16 peak
6.666* TFLOPS
Scope and conditions
Arithmetic density
Not stated*
Reported theoretical FP32 peak
3.333* TFLOPS
Scope and conditions
Arithmetic density
Not stated*
Declared L1 cache configuration
128 KB (per SM)*
Declared GPU L2 cache
6 MB*
Reported display outputs
Portable Device Dependent*
Theoretical pixel fill rate
22.32 GPixel/s*
Reported chip die area
200 mm²*
Chip foundry
Samsung*
Process node declaration
8 nm*
Graphics processor
GA10B*
Reported video memory
32* GB
Rated memory bandwidth
204.8 GB/s*
Video memory type
LPDDR5*
Raster output units
24*
Reported shading units
1,792*
Card or module format
IGP*
Streaming multiprocessors
14*
2:4 tensor sparsity
Yes*
Declared GPU TDP
40 W*
Tensor cores
56*
Theoretical texture fill rate
52.08 GTexel/s*
Texture mapping units
56*
GPU cores
1,792*
Video memory
32* GB
Video memory type
LPDDR5*

* Provisional data