AMD Instinct MI250X vs NVIDIA GRID M40

AMD Instinct MI250X
NVIDIA GRID M40
14080 Shaders
128GB HBM2e
1700MHz
384 Shaders
8GB GDDR5
1033MHz
Peak AI Performance
382.98 TOPS
INT4 Tensor
Peak AI Performance
3.17 TOPS
INT8
FP32
47.87 TFLOPS
FP32
790 GFLOPS
FP16
95.74 TFLOPS
-
-
Form Factor
OAM Module
-
Form Factor
PCIe Card
2.0-Slots
TDP
560W
TDP
50W
-
-
-
-
-
-
-
-
-
-
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
382.98 TOPS
INT4 Tensor
Peak AI Performance
3.17 TOPS
INT8
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
FP16
95.74 TFLOPS
382.98 TFLOPS Tensor (FP16 Accumulate)
-
382.98 TFLOPS Tensor (FP32 Accumulate)
-
-
-
-
-
-
-
FP32
47.87 TFLOPS
95.74 TFLOPS Tensor
-
FP32
790 GFLOPS
-
-
FP64
47.87 TFLOPS
95.74 TFLOPS Tensor
FP64
30 GFLOPS
-
BF16
-
382.98 TFLOPS Tensor
-
-
-
-
-
-
-
-
-
-
-
INT4
382.98 TOPS Tensor
-
-
-
-
INT8
-
382.98 TOPS Tensor
-
INT8
3.17 TOPS
-
-
-
-
INT32
0.79 TOPS
-
-
-
-
Pixel Fillrate
-
Pixel Fillrate
16.528 GPixel/s
-
-
-
-
Texture Fillrate
1496 GTexel/s
Texture Fillrate
33.056 GTexel/s
Manufacturer
AMD
Manufacturer
NVIDIA
Chip Designer
AMD
Chip Designer
NVIDIA
Architecture
CDNA 2
Architecture
Maxwell
Family
Instinct
Family
GRID
Codename
Aldebaran
Aldebaran XTX
Variant
Aldebaran XTX
Codename
NV117
GM107
-
-
Market Segment
Server
Market Segment
Server
Release Date
11/8/2021
Release Date
5/18/2016
Foundry
TSMC
-
Foundry
TSMC
-
Fabrication Node
N6
-
Fabrication Node
28nm
-
Die Size
2x 724 mm²
-
Die Size
148 mm²
-
Transistor Count
2x 28 Billion
-
Transistor Count
1.9 Billion
-
Transistor Density
38.67M/mm²
-
Transistor Density
12.64M/mm²
-
Form
OAM Module
Form
PCIe Card
Shading Units
14080 Shaders
-
Shading Units
384 Shaders
-
Texture Mapping Units
880 TMUs
Texture Mapping Units
32 TMUs
Render Output Units
-
Render Output Units
16 ROPs
Tensor Cores
880 T-Cores
-
-
-
-
-
-
-
-
Streaming Multiprocessors
3 SMs
Compute Units
220 CUs
-
-
-
-
-
-
-
-
-
-
-
-
1000MHz Base
1700MHz
-
-
-
1033MHz
-
-
-
-
L1
-
-
16KB/CU
-
L1
-
64KB/SM
-
-
L2
16MB Shared
L2
2MB Shared
-
-
-
-
-
-
128GB
HBM2e
ECC
8GB
GDDR5
-
Bus Width
8192Bit
Bus Width
128Bit
Clock
1600MHz
Transfer Rate
3.2GT/s
Bandwidth
3276.8GB/s
Clock
1300MHz
Transfer Rate
5.2GT/s
Bandwidth
83.2GB/s
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
TDP
560W
TDP
50W
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
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
-
-
-
-
Model
2x VCN 2.6
Model
NVENC 2
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
-
-
-
-
-
No Decoders
Model
NVDEC 1
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
-
-
-
AVC (H.264)
-
-
-
-
-
-
-
-
-
Direct X
12
Direct 3D
11_0
-
-
OpenCL
3.0
-
-
OpenGL
4.6
OpenCL
3.0
Vulkan
1.3
-
-
-
-
GFX
9.4
-
-
-
-
Shader Model
6.7
CUDA
5.0
-
-
PureVideo HD
VP6
VDPAU
Feature Set E
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
PCIe Version
4.0
PCIe Lanes
16
Slots Required
2.0
PCIe Version
3.0
PCIe Lanes
16
Multi GPU Support
Supported
Type
Infinity Fabric
-
-
-
-
-
-
-
-
-
-
Height
111 mm (4.37 in)
Width
267 mm (10.51 in)
Depth
40 mm (1.57 in)
AMD Instinct MI250X AMD Instinct MI250
AMD Instinct MI250X vs AMD Instinct MI250
Change Comparison