Intel Gaudi vs NVIDIA T4
Intel Gaudi
NVIDIA T4
|
2048 Shaders
32GB HBM2
610MHz
|
2560 Shaders
16GB GDDR6
1590MHz
|
|
Peak AI Performance
159.91 TFLOPS
BF16 Tensor
|
Peak AI Performance
260.51 TOPS
INT4 Tensor
|
|
-
-
|
FP32
8.14 TFLOPS
|
|
-
-
|
FP16
16.28 TFLOPS
|
|
Form Factor
OAM Module
-
|
Form Factor
PCIe Card
1.0-Slots
|
|
TDP
350W
|
TDP
70W
|
|
-
-
-
-
-
|
-
-
-
-
-
|
|
GB6 OpenCL N/A
0%
|
GB6 OpenCL N/A
0%
|
|
GB6 Metal N/A
0%
|
GB6 Metal N/A
0%
|
|
GB6 Vulkan N/A
0%
|
GB6 Vulkan N/A
0%
|
|
GB5 OpenCL N/A
0%
|
GB5 OpenCL N/A
0%
|
|
GB5 CUDA N/A
0%
|
GB5 CUDA N/A
0%
|
|
GB5 Metal N/A
0%
|
GB5 Metal N/A
0%
|
|
GB5 Vulkan N/A
0%
|
GB5 Vulkan N/A
0%
|
|
OCT 2020.1 N/A
0%
|
OCT 2020.1 N/A
0%
|
|
OCT Metal N/A
0%
|
OCT Metal N/A
0%
|
|
Peak AI
Performance
159.91 TFLOPS
BF16 Tensor
|
Peak AI
Performance
260.51 TOPS
INT4 Tensor
|
|
-
-
-
|
-
-
-
|
|
-
-
-
-
-
-
|
-
-
-
-
-
-
|
|
-
-
-
-
-
-
|
FP16
16.28 TFLOPS
65.13 TFLOPS
Tensor (FP16 Accumulate)
-
65.13 TFLOPS
Tensor (FP32 Accumulate)
-
|
|
-
-
-
-
|
FP32
8.14 TFLOPS
-
-
|
|
-
-
-
|
FP64
250 GFLOPS
-
|
|
BF16
-
159.91 TFLOPS
Tensor
-
|
-
-
-
-
|
|
-
-
-
|
-
-
-
|
|
-
-
-
|
INT4
260.51 TOPS
Tensor
-
|
|
-
-
-
-
|
INT8
-
130.25 TOPS
Tensor
-
|
|
-
-
|
-
-
|
|
-
-
|
Ray Tracing
31.8 TOPS
|
|
Pixel
Fillrate
-
|
Pixel
Fillrate
101.76 GPixel/s
|
|
-
-
|
-
-
|
|
Texture
Fillrate
-
|
Texture
Fillrate
254.4 GTexel/s
|
|
Manufacturer
Intel
|
Manufacturer
NVIDIA
|
|
Chip Designer
Intel
|
Chip Designer
NVIDIA
|
|
Architecture
Gaudi
|
Architecture
Turing
|
|
Family
Gaudi
|
Family
Server
|
|
Codename
Gaudi
HL-2000
Variant
HL-2000
|
Codename
NV164
TU104
Variant
TU104-895-A1
|
|
Market Segment
Server
|
Market Segment
Server
|
|
Release Date
6/17/2019
|
Release Date
9/13/2018
|
|
Foundry
TSMC
-
|
Foundry
TSMC
-
|
|
Fabrication Node
16FF
-
|
Fabrication Node
12FFN
-
|
|
-
-
-
|
Die Size
545 mm²
-
|
|
-
-
|
Transistor Count
13.6 Billion
-
|
|
-
-
-
|
Transistor Density
24.95M/mm²
-
|
|
Form
OAM Module
|
Form
PCIe Card
|
|
Shading Units
2048 Shaders
-
|
Shading Units
2560 Shaders
-
|
|
Texture Mapping Units
-
|
Texture Mapping Units
160 TMUs
|
|
Render Output Units
-
|
Render Output Units
64 ROPs
|
|
Tensor Cores
8 T-Cores
|
Tensor Cores
320 T-Cores
|
|
-
-
|
Ray-Tracing Cores
40 RT-Cores
|
|
-
-
|
Streaming Multiprocessors
40 SMs
|
|
Compute Units
1 CU
|
-
-
|
|
-
-
|
-
-
|
|
-
-
|
-
-
|
|
-
-
-
610MHz
|
-
-
585MHz Base
1590MHz
|
|
-
-
|
-
-
|
|
L1
-
-
-
Unknown
|
L1
32KB/SM Tex
64KB/SM
-
-
|
|
L2
Unknown
|
L2
4MB Shared
|
|
-
-
-
|
-
-
-
|
|
32GB
HBM2
ECC
|
16GB
GDDR6
-
|
|
Bus Width
4096Bit
|
Bus Width
256Bit
|
|
Clock
980MHz
Transfer Rate
2GT/s
Bandwidth
1003.5GB/s
|
Clock
625MHz
Transfer Rate
5GT/s
Bandwidth
160GB/s
|
|
-
-
-
-
eSRAM
24MB
3200GB/s
-
-
|
-
-
-
-
-
-
-
-
-
|
|
TDP
350W
|
TDP
70W
|
|
-
-
|
-
-
|
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
|
|
Max Resolution
Unknown
|
Max Resolution
Unknown
|
|
Max Resolution Refresh Rate
-
|
Max Resolution Refresh Rate
-
|
|
Variable Refresh Rate
-
-
-
|
Variable Refresh Rate
G-Sync
FreeSync
-
|
|
Display Stream Compression (DSC)
Not Supported
|
Display Stream Compression (DSC)
Not Supported
|
|
Multi Monitor Support
Unknown
|
Multi Monitor Support
Unknown
|
|
-
-
|
-
-
|
|
No Encoders
-
|
Model
NVENC 7
|
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
|
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
|
|
No Decoders
|
Model
NVDEC 4
|
|
-
-
-
-
-
-
-
-
-
-
-
-
-
-
|
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
|
|
-
-
-
-
|
Direct X
12
Direct 3D
12_2
|
|
-
-
OpenCL
3.0
-
-
|
OpenGL
4.6
OpenCL
3.0
Vulkan
1.2
|
|
-
-
-
-
-
-
-
-
-
-
|
Shader Model
6.6
CUDA
7.5
-
-
PureVideo HD
VP10
VDPAU
Feature Set J
|
|
Not a Card
-
-
-
|
-
-
-
-
|
|
-
-
-
-
-
-
-
-
|
-
-
-
-
-
-
-
-
|
|
-
-
PCIe Version
4.0
PCIe Lanes
16
|
Slots Required
1.0
PCIe Version
3.0
PCIe Lanes
16
|
|
Multi GPU Support
Supported
Type
RoCE
|
Multi GPU Support
Supported
Type
NVLink
|
|
-
-
-
-
-
-
|
Height
69 mm (2.72 in)
Width
168 mm (6.61 in)
Depth
20 mm (0.79 in)
|
Copy Link