AMD Radeon Instinct MI100 vs NVIDIA GRID K2
We compared two Professional market GPUs: 32GB VRAM Radeon Instinct MI100 and 4GB VRAM GRID K2 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
AMD Radeon Instinct MI100 's Advantages
Released 7 years and 6 months late
Boost Clock1502MHz
More VRAM (32GB vs 4GB)
Larger VRAM bandwidth (1229GB/s vs 160.0GB/s)
6144 additional rendering cores
NVIDIA GRID K2 's Advantages
Lower TDP (225W vs 300W)
Score
Benchmark
FP32 (float)
Radeon Instinct MI100
+907%
23.07 TFLOPS
GRID K2
2.289 TFLOPS
Graphics Card
Nov 2020
Release Date
May 2013
Radeon Instinct
Generation
GRID
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 3.0 x16
Clock Speeds
1000 MHz
Base Clock
-
1502 MHz
Boost Clock
-
1200 MHz
Memory Clock
1250 MHz
Memory
32GB
Memory Size
4GB
HBM2
Memory Type
GDDR5
4096bit
Memory Bus
256bit
1229GB/s
Bandwidth
160.0GB/s
Render Config
120
Compute Units
-
-
-
-
7680
Shading Units
1536
480
TMUs
128
64
ROPs
32
-
-
-
-
-
-
16 KB (per CU)
L1 Cache
16 KB (per SMX)
8 MB
L2 Cache
512 KB
-
-
-
Theoretical Performance
96.13 GPixel/s
Pixel Rate
23.84 GPixel/s
721.0 GTexel/s
Texture Rate
95.36 GTexel/s
184.6 TFLOPS
FP16 (half)
-
23.07 TFLOPS
FP32 (float)
2.289 TFLOPS
11.54 TFLOPS
FP64 (double)
95.36 GFLOPS
Board Design
300W
TDP
225W
700 W
Suggested PSU
550 W
No outputs
Outputs
No outputs
2x 8-pin
Power Connectors
1x 6-pin + 1x 8-pin
Graphics Processor
Arcturus
GPU Name
GK104
Arcturus XL
GPU Variant
GK104-895-A2
CDNA 1.0
Architecture
Kepler
TSMC
Foundry
TSMC
7 nm
Process Size
28 nm
25.6 billion
Transistors
3.54 billion
750 mm²
Die Size
294 mm²
Graphics Features
N/A
DirectX
12 (11_0)
N/A
OpenGL
4.6
2.1
OpenCL
3.0
N/A
Vulkan
1.2.175
-
CUDA
3.0
N/A
Shader Model
6.5 (5.1)
