Report an Error

NVIDIA Jetson AGX Xavier

Graphics Processor
GV10B
Cores
512
TMUs
32
ROPs
16
Memory Size
System Shared
Memory Type
System Shared
Bus Width
System Shared
The Jetson AGX Xavier is a mobile integrated graphics solution by NVIDIA, launched in October 2018. Built on the 12 nm process, and based on the GV10B graphics processor, the device supports DirectX 12.0. It features 512 shading units, 32 texture mapping units and 16 ROPs. Also included are 64 tensor cores which help improve the speed of machine learning applications. The GPU is operating at a frequency of 854 MHz, which can be boosted up to 1377 MHz.
Its power draw is rated at 30 W maximum.

Graphics Processor

GPU Name
GV10B
Architecture
Volta
Foundry
TSMC
Process Size
12 nm
Transistors
9,000 million
Die Size
350 mm²

Integrated Graphics

Release Date
Oct 2018
Generation
Tegra
Production
Active
Bus Interface
IGP

Relative Performance

6%
GeForce 210
8%
GeForce 9400 GT
10%
Radeon HD 4550
13%
Radeon HD 5450
15%
Radeon HD 6450
16%
GeForce GT 520
20%
GeForce GT 220
26%
GeForce GT 430
27%
Radeon HD 5570
29%
Radeon HD 4670
30%
GeForce GT 440
31%
GeForce GT 240
36%
GeForce 9600 GT
38%
Radeon HD 5670
41%
GeForce GT 640
42%
GeForce 9800 GT
44%
Radeon HD 6670
45%
Radeon HD 4830
48%
Radeon HD 4770
51%
Radeon HD 4850
52%
GeForce GTS 250
53%
GeForce GTS 450
55%
Radeon HD 5750
63%
Radeon HD 7750
64%
Radeon HD 5770
65%
GeForce GTX 260
66%
Radeon HD 4870
66%
Radeon Vega 8
66%
GeForce GTX 550 Ti
67%
GeForce GTX 650
71%
Radeon HD 5830
75%
Radeon HD 6790
76%
Radeon HD 4890
79%
GeForce GTX 460
79%
Radeon RX Vega 11
80%
GeForce GT 1030
81%
GeForce GTX 275
81%
Radeon HD 7770 GHz Edition
81%
GeForce GTX 280
83%
GeForce GTX 465
87%
Radeon HD 6850
88%
GeForce GTX 285
95%
Radeon HD 5850
96%
GeForce GTX 650 Ti
96%
Radeon HD 7790
100%
Jetson AGX Xavier
101%
Radeon RX 550
101%
GeForce GTX 470
104%
Radeon HD 6870
111%
Radeon HD 5870
115%
GeForce GTX 560 Ti
116%
Radeon HD 4870 X2
118%
GeForce GTX 750 Ti
118%
Radeon HD 6950
122%
GeForce GTX 295
124%
GeForce GTX 650 Ti Boost
125%
Radeon HD 7850
127%
GeForce GTX 480
133%
Radeon HD 6970
135%
Radeon R7 265
138%
Radeon RX 460
138%
Radeon R7 370
138%
GeForce GTX 660
138%
GeForce GTX 570
141%
Radeon RX 560
150%
GeForce GTX 950
151%
Radeon HD 7870 GHz Edition
154%
Radeon R9 270X
156%
GeForce GTX 660 Ti
157%
GeForce GTX 580
158%
GeForce GTX 1050
159%
Radeon HD 7950
162%
Radeon HD 5970
166%
GeForce GTX 760
179%
GeForce GTX 670
184%
GeForce GTX 960
188%
Radeon R9 380
190%
Radeon R9 285
191%
Radeon HD 7970
193%
GeForce GTX 680
198%
GeForce GTX 1050 Ti
203%
GeForce GTX 770
208%
Radeon HD 6990
213%
Radeon R9 280X
216%
GeForce GTX 590
216%
Radeon HD 7970 GHz Edition
239%
GeForce GTX 780
248%
GeForce GTX 1650
263%
Radeon RX 470
267%
Radeon R9 290
281%
Radeon RX 570
285%
Radeon R9 390
286%
Radeon R9 290X
290%
GeForce GTX 970
294%
GeForce GTX TITAN
298%
GeForce GTX 780 Ti
302%
Radeon R9 390X
303%
Radeon HD 7990
305%
Radeon RX 480
307%
GeForce GTX 690
324%
Radeon RX 580
331%
GeForce GTX 980
335%
Radeon R9 FURY
335%
GeForce GTX 1060 6 GB
350%
Radeon RX 590
368%
Radeon R9 295X2
370%
GeForce GTX 1660
375%
Radeon R9 FURY X
381%
GeForce GTX 980 Ti
391%
GeForce GTX TITAN X
425%
GeForce GTX 1660 Ti
425%
GeForce GTX 1070
453%
Radeon RX Vega 56
480%
GeForce GTX 1070 Ti
494%
Radeon RX Vega 64
501%
GeForce GTX 1080
501%
GeForce RTX 2060
521%
Radeon RX 5700
563%
GeForce RTX 2060 SUPER
583%
GeForce RTX 2070
597%
Radeon RX 5700 XT
604%
Radeon VII
619%
TITAN X Pascal
638%
GeForce GTX 1080 Ti
645%
GeForce RTX 2070 SUPER
686%
GeForce RTX 2080
727%
GeForce RTX 2080 SUPER
803%
GeForce RTX 2080 Ti
Based on TPU review data: "Performance Summary" at 1920x1080
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

Base Clock
854 MHz
Boost Clock
1377 MHz
Memory Clock
System Shared

Memory

Memory Size
System Shared
Memory Type
System Shared
Memory Bus
System Shared
Bandwidth
System Dependent

Render Config

Shading Units
512
TMUs
32
ROPs
16
SM Count
8
Tensor Cores
64
L1 Cache
128 KB (per SM)
L2 Cache
512 KB

Theoretical Performance

Pixel Rate
22.03 GPixel/s
Texture Rate
44.06 GTexel/s
FP16 (half) performance
2.820 TFLOPS (2:1)
FP32 (float) performance
1,410 GFLOPS
FP64 (double) performance
705.0 GFLOPS (1:2)

Board Design

Slot Width
IGP
TDP
30 W
Outputs
No outputs

Graphics Features

DirectX
12.0 (12_1)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.1.122
CUDA
7.2
Shader Model
6.4