NVIDIA A16 vs AMD Radeon Instinct MI25

NVIDIA A16
AMD Radeon Instinct MI25
4x 1280 Shaders
64GB (4x 16GB) GDDR6
1695MHz
4096 Shaders
16GB HBM2
1500MHz
Peak AI Performance
555.42 TOPS
INT4 Tensor Sparse
Peak AI Performance
24.58 TFLOPS
FP16
FP32
17.36 TFLOPS
FP32
12.29 TFLOPS
FP16
17.36 TFLOPS
FP16
24.58 TFLOPS
Form Factor
PCIe Card
2.0-Slots
Form Factor
PCIe Card
2.0-Slots
TDP
250W
TDP
300W
Power Connectors
-
-
-
-
Power Connectors
-
2x 8-Pin
-
-
GB6 OpenCL N/A
0%
GB6 Metal N/A
0%
GB6 Metal 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
555.42 TOPS
INT4 Tensor Sparse
Peak AI Performance
24.58 TFLOPS
FP16
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
FP16
17.36 TFLOPS
69.43 TFLOPS Tensor (FP16 Accumulate)
138.86 TFLOPS Tensor (FP16 Accumulate) Sparse
69.43 TFLOPS Tensor (FP32 Accumulate)
138.86 TFLOPS Tensor (FP32 Accumulate) Sparse
FP16
24.58 TFLOPS
-
-
-
-
FP32
17.36 TFLOPS
-
-
FP32
12.29 TFLOPS
-
-
FP64
270 GFLOPS
-
FP64
770 GFLOPS
-
BF16
17.36 TFLOPS
69.43 TFLOPS Tensor
138.86 TFLOPS Tensor Sparse
-
-
-
-
TF32
34.71 TFLOPS Tensor
69.42 TFLOPS Tensor Sparse
-
-
-
INT4
277.71 TOPS Tensor
555.42 TOPS Tensor Sparse
-
-
-
INT8
-
138.86 TOPS Tensor
277.71 TOPS Tensor Sparse
-
-
-
-
INT32
8.68 TOPS
-
-
Ray Tracing
8.5 TOPS
-
-
Pixel Fillrate
54.24 GPixel/s
Pixel Fillrate
96 GPixel/s
-
-
-
-
Texture Fillrate
67.8 GTexel/s
Texture Fillrate
384 GTexel/s
Manufacturer
NVIDIA
Manufacturer
AMD
Chip Designer
NVIDIA
Chip Designer
AMD
Architecture
Ampere
Architecture
GCN 5
Family
Server
Family
Instinct
Codename
NV177
GA107
-
-
Codename
Greenland
Vega 10
Variant
Vega 10 XT GL
Market Segment
Server
Market Segment
Server
Release Date
4/12/2021
Release Date
6/20/2017
Foundry
Samsung
-
Foundry
GlobalFoundries
-
Fabrication Node
8N
-
Fabrication Node
14LPP
-
Die Size
4x 200 mm²
-
Die Size
486 mm²
-
Transistor Count
4x 8.7 Billion
-
Transistor Count
12.5 Billion
-
Transistor Density
43.50M/mm²
-
Transistor Density
25.72M/mm²
-
Form
PCIe Card
Form
PCIe Card
Shading Units
4x 1280 Shaders
-
Shading Units
4096 Shaders
-
Texture Mapping Units
4x 40 TMUs
Texture Mapping Units
256 TMUs
Render Output Units
4x 32 ROPs
Render Output Units
64 ROPs
Tensor Cores
4x 40 T-Cores
-
-
Ray-Tracing Cores
4x 10 RT-Cores
-
-
Streaming Multiprocessors
4x 10 SMs
-
-
-
-
Compute Units
64 CUs
-
-
-
-
-
-
-
-
-
-
885MHz Base
1695MHz
-
-
1400MHz Base
1500MHz
-
-
-
-
L1
64KB/SM Tex
128KB/SM
-
-
L1
-
-
16KB/CU
-
L2
2MB Shared
L2
4MB Shared
-
-
-
-
-
-
64GB (4x 16GB)
GDDR6
-
16GB
HBM2
ECC
Bus Width
128Bit
Bus Width
2048Bit
Clock
905MHz
Transfer Rate
7.2GT/s
Bandwidth
115.9GB/s
Clock
945MHz
Transfer Rate
1.9GT/s
Bandwidth
483.8GB/s
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
TDP
250W
TDP
300W
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
No Ports
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
1x mini-DisplayPort 1.4
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Max Resolution
Unknown
Max Resolution
7680x4320
Max Resolution Refresh Rate
-
Max Resolution Refresh Rate
60Hz
Variable Refresh Rate
G-Sync
FreeSync
-
Variable Refresh Rate
-
FreeSync
-
Display Stream Compression (DSC)
Not Supported
Display Stream Compression (DSC)
Not Supported
Multi Monitor Support
Unknown
Multi Monitor Support
3
-
-
Content Protection
HDCP 2.2
Model
NVENC 7
Model
VCE 4.0
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
Codec
-
-
-
-
-
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
Model
NVDEC 5
Model
UVD 7.0
Codec
MPEG-1
MPEG-2
MPEG-4
-
VC-1
VP8
VP9
-
AVC (H.264)
HEVC (H.265)
-
AV1
-
-
Codec
MPEG-1
MPEG-2
MPEG-4
JPEG
VC-1
-
-
-
AVC (H.264)
HEVC (H.265)
-
-
-
-
Direct X
12
Direct 3D
12_2
Direct X
12
Direct 3D
12_1
OpenGL
4.6
OpenCL
3.0
Vulkan
1.2
OpenGL
4.6
OpenCL
2.1
Vulkan
1.3
Shader Model
6.6
CUDA
8.6
-
-
PureVideo HD
VP11
VDPAU
Feature Set K
Shader Model
6.7
-
-
GFX
9
-
-
-
-
-
-
-
-
-
-
-
-
Power Connectors
-
-
-
-
1x 8-Pin EPS
-
-
Power Connectors
-
-
-
2x 8-Pin
-
-
-
Slots Required
2.0
PCIe Version
4.0
PCIe Lanes
8
Slots Required
2.0
PCIe Version
3.0
PCIe Lanes
16
-
-
-
-
Multi GPU Support
Supported
Type
CrossFire XDMA
Height
112 mm (4.41 in)
Width
267 mm (10.51 in)
Depth
40 mm (1.57 in)
Height
111 mm (4.37 in)
Width
267 mm (10.51 in)
Depth
37 mm (1.46 in)
Change Comparison