NVIDIA Tesla K20Xm vs AMD Radeon Instinct MI50
We compared two Professional market GPUs: 6GB VRAM Tesla K20Xm and 8GB VRAM Radeon Instinct MI50 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA Tesla K20Xm 's Advantages
640 additional rendering cores
AMD Radeon Instinct MI50 's Advantages
Released 12 years and 6 months late
Boost Clock3230MHz
More VRAM (8GB vs 6GB)
Larger VRAM bandwidth (322.3GB/s vs 249.6GB/s)
Lower TDP (150W vs 235W)
Score
Benchmark
FP32 (float)
Tesla K20Xm
3.935 TFLOPS
Radeon Instinct MI50
+572%
26.46 TFLOPS
Graphics Card
Nov 2012
Release Date
May 2025
Tesla
Generation
Navi IV
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 5.0 x8
Clock Speeds
-
Base Clock
2220 MHz
-
Boost Clock
3230 MHz
1300 MHz
Memory Clock
2518 MHz
Memory
6GB
Memory Size
8GB
GDDR5
Memory Type
GDDR6
384bit
Memory Bus
128bit
249.6GB/s
Bandwidth
322.3GB/s
Render Config
-
Compute Units
32
-
-
-
2688
Shading Units
2048
224
TMUs
128
48
ROPs
64
-
-
-
-
RT Cores
32
16 KB (per SMX)
L1 Cache
16 KB (per CU)
1536 KB
L2 Cache
4 MB
-
L3 Cache
32 MB
Theoretical Performance
40.99 GPixel/s
Pixel Rate
206.7 GPixel/s
164.0 GTexel/s
Texture Rate
413.4 GTexel/s
-
FP16 (half)
52.92 TFLOPS
3.935 TFLOPS
FP32 (float)
26.46 TFLOPS
1312 GFLOPS
FP64 (double)
826.9 GFLOPS
Board Design
235W
TDP
150W
550 W
Suggested PSU
450 W
No outputs
Outputs
1x HDMI 2.1b
3x DisplayPort 2.1a
-
Power Connectors
1x 8-pin
Graphics Processor
GK110
GPU Name
Navi 44
-
GPU Variant
Navi 44 XT
Kepler
Architecture
RDNA 4.0
TSMC
Foundry
TSMC
28 nm
Process Size
4 nm
7.08 billion
Transistors
Unknown
561 mm²
Die Size
153 mm²
Graphics Features
12 (11_0)
DirectX
12 Ultimate (12_2)
4.6
OpenGL
4.6
3.0
OpenCL
2.2
1.1
Vulkan
1.3
3.5
CUDA
-
5.1
Shader Model
6.8
