Report an Error

NVIDIA Tesla T4

Graphics Processor
TU104
Cores
2560
TMUs
160
ROPs
64
Memory Size
16 GB
Memory Type
GDDR6
Bus Width
256 bit
The Tesla T4 is a professional graphics card by NVIDIA, launched in September 2018. Built on the 12 nm process, and based on the TU104 graphics processor, the card supports DirectX 12.0. The TU104 graphics processor is a large chip with a die area of 545 mm² and 13,600 million transistors. Unlike the fully unlocked GeForce RTX 2080, which uses the same GPU but has all 2944 shaders enabled, NVIDIA has disabled some shading units on the Tesla T4 to reach the product's target shader count. It features 2560 shading units, 160 texture mapping units and 64 ROPs. Also included are 320 tensor cores which help improve the speed of machine learning applications. The card also has 40 raytracing acceleration cores. NVIDIA has placed 16,384 MB GDDR6 memory on the card, which are connected using a 256-bit memory interface. The GPU is operating at a frequency of 585 MHz, which can be boosted up to 1590 MHz, memory is running at 1250 MHz.
Being a single-slot card, the NVIDIA Tesla T4 does not require any additional power connector, its power draw is rated at 75 W maximum. This device has no display connectivity, as it is not designed to have monitors connected to it. Tesla T4 is connected to the rest of the system using a PCI-Express 3.0 x16 interface.

Graphics Processor

GPU Name
TU104
Architecture
Turing
Foundry
TSMC
Process Size
12 nm
Transistors
13,600 million
Die Size
545 mm²

Graphics Card

Release Date
Sep 13th, 2018
Generation
Tesla
(Txx)
Production
Active
Bus Interface
PCIe 3.0 x16

Relative Performance

1%
GeForce 210
1%
GeForce 9400 GT
2%
Radeon HD 4550
2%
Radeon HD 5450
2%
Radeon HD 6450
3%
GeForce GT 520
3%
GeForce GT 220
4%
GeForce GT 430
4%
Radeon HD 5570
4%
Radeon HD 4670
5%
GeForce GT 440
5%
GeForce GT 240
6%
GeForce 9600 GT
6%
Radeon HD 5670
6%
GeForce GT 640
7%
GeForce 9800 GT
7%
Radeon HD 6670
7%
Radeon HD 4830
7%
Radeon HD 4770
8%
Radeon HD 4850
8%
GeForce GTS 250
8%
GeForce GTS 450
9%
Radeon HD 5750
10%
Radeon HD 7750
10%
Radeon HD 5770
10%
GeForce GTX 260
10%
Radeon HD 4870
10%
Radeon Vega 8
10%
GeForce GTX 550 Ti
10%
GeForce GTX 650
11%
Radeon HD 5830
12%
Radeon HD 6790
12%
Radeon HD 4890
12%
GeForce GTX 460
12%
Radeon RX Vega 11
12%
GeForce GT 1030
13%
GeForce GTX 275
13%
Radeon HD 7770 GHz Edition
13%
GeForce GTX 280
13%
GeForce GTX 465
13%
Radeon HD 6850
14%
GeForce GTX 285
15%
Radeon HD 5850
15%
Radeon HD 7790
15%
GeForce GTX 650 Ti
16%
Radeon RX 550
16%
GeForce GTX 470
16%
Radeon HD 6870
17%
Radeon HD 5870
18%
GeForce GTX 560 Ti
18%
Radeon HD 4870 X2
18%
GeForce GTX 750 Ti
18%
Radeon HD 6950
19%
GeForce GTX 295
19%
GeForce GTX 650 Ti Boost
19%
Radeon HD 7850
20%
GeForce GTX 480
21%
Radeon HD 6970
21%
Radeon R7 265
21%
Radeon R7 370
21%
Radeon RX 560
21%
GeForce GTX 660
21%
GeForce GTX 570
23%
GeForce GTX 950
23%
Radeon HD 7870 GHz Edition
24%
Radeon R9 270X
24%
GeForce GTX 660 Ti
24%
GeForce GTX 580
25%
GeForce GTX 1050
25%
Radeon HD 7950
25%
Radeon RX 460
25%
Radeon HD 5970
26%
GeForce GTX 760
28%
GeForce GTX 670
29%
GeForce GTX 960
29%
Radeon R9 380
29%
Radeon R9 285
30%
Radeon HD 7970
30%
GeForce GTX 680
31%
GeForce GTX 1050 Ti
32%
GeForce GTX 770
32%
Radeon HD 6990
33%
Radeon R9 280X
33%
GeForce GTX 590
34%
Radeon HD 7970 GHz Edition
37%
GeForce GTX 780
41%
Radeon RX 470
41%
Radeon R9 290
44%
Radeon R9 390
44%
Radeon R9 290X
45%
GeForce GTX 970
46%
GeForce GTX TITAN
46%
Radeon RX 570
46%
GeForce GTX 780 Ti
47%
Radeon R9 390X
47%
Radeon HD 7990
47%
Radeon RX 480
48%
GeForce GTX 690
50%
Radeon RX 580
51%
GeForce GTX 980
52%
Radeon R9 FURY
52%
GeForce GTX 1060 6 GB
57%
Radeon R9 295X2
58%
Radeon R9 FURY X
59%
GeForce GTX 980 Ti
64%
GeForce GTX TITAN X
70%
GeForce GTX 1070
72%
Radeon RX Vega 56
78%
GeForce GTX 1070 Ti
78%
Radeon RX Vega 64
82%
GeForce GTX 1080
92%
GeForce RTX 2070
96%
TITAN X Pascal
99%
GeForce GTX 1080 Ti
100%
Tesla T4
107%
GeForce RTX 2080
122%
GeForce RTX 2080 Ti
Based on TPU review data: "Performance Summary" at 1920x1080
Performance estimated based on architecture, shader count and clocks.

Clock Speeds

GPU Clock
585 MHz
Boost Clock
1590 MHz
Memory Clock
1250 MHz
10000 MHz effective

Memory

Memory Size
16 GB
Memory Type
GDDR6
Memory Bus
256 bit
Bandwidth
320.0 GB/s

Render Config

Shading Units
2560
TMUs
160
ROPs
64
SM Count
40
Tensor Cores
320
RT Cores
40

Theoretical Performance

Pixel Rate
101.8 GPixel/s
Texture Rate
254.4 GTexel/s
FP16 (half) performance
65,126 GFLOPS (8:1)
FP32 (float) performance
8,141 GFLOPS
FP64 (double) performance
254.4 GFLOPS (1:32)

Board Design

Slot Width
Single-slot
TDP
75 W
Outputs
No outputs
Power Connectors
None

Graphics Features

DirectX
12.0 (12_1)
OpenGL
4.6
OpenCL
1.2
Vulkan
1.1.95
CUDA
7.5
Shader Model
6.3

Card Notes

Some Specs Unknown