AMD Instinct MI300A vs NVIDIA Tesla K8
We compared two Professional market GPUs: 128GB VRAM AMD Instinct MI300A and 8GB VRAM Tesla K8 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
AMD Instinct MI300A 's Advantages
Released 9 years and 3 months late
Boost Clock has increased by 159% (2100MHz vs 811MHz)
More VRAM (128GB vs 8GB)
Larger VRAM bandwidth (5300GB/s vs 160.0GB/s)
13056 additional rendering cores
NVIDIA Tesla K8 's Advantages
Lower TDP (100W vs 760W)
Score
Benchmark
FP32 (float)
AMD Instinct MI300A
+4821%
122.6 TFLOPS
Tesla K8
2.491 TFLOPS
Graphics Card
Dec 2023
Release Date
Sep 2014
Instinct
Generation
Tesla
Professional
Type
Professional
PCIe 5.0 x16
Bus Interface
PCIe 2.0 x16
Clock Speeds
1000 MHz
Base Clock
693 MHz
2100 MHz
Boost Clock
811 MHz
5200 MHz
Memory Clock
1250 MHz
Memory
128GB
Memory Size
8GB
HBM3
Memory Type
GDDR5
8192bit
Memory Bus
256bit
5300GB/s
Bandwidth
160.0GB/s
Render Config
228
Compute Units
-
-
-
-
14592
Shading Units
1536
880
TMUs
128
0
ROPs
32
-
-
-
-
-
-
16 KB (per CU)
L1 Cache
16 KB (per SMX)
16 MB
L2 Cache
512 KB
-
-
-
Theoretical Performance
0 MPixel/s
Pixel Rate
25.95 GPixel/s
1496 GTexel/s
Texture Rate
103.8 GTexel/s
980.6 TFLOPS
FP16 (half)
-
122.6 TFLOPS
FP32 (float)
2.491 TFLOPS
61.3 TFLOPS
FP64 (double)
103.8 GFLOPS
Board Design
760W
TDP
100W
1000 W
Suggested PSU
300 W
No outputs
Outputs
No outputs
None
Power Connectors
1x 6-pin
Graphics Processor
MI300
GPU Name
GK104
-
-
-
CDNA 3.0
Architecture
Kepler
TSMC
Foundry
TSMC
5 nm
Process Size
28 nm
146 billion
Transistors
3.54 billion
1017 mm²
Die Size
294 mm²
Graphics Features
N/A
DirectX
12 (11_0)
N/A
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.1
-
CUDA
3.0
N/A
Shader Model
5.1
