GRID M40 vs Radeon Instinct MI300
We compared two Professional market GPUs: 8GB VRAM GRID M40 and 128GB VRAM Radeon Instinct MI300 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
GRID M40 's Advantages
Lower TDP (50W vs 600W)
Radeon Instinct MI300 's Advantages
Released 6 years and 8 months late
Boost Clock1700MHz
More VRAM (128GB vs 8GB)
Larger VRAM bandwidth (3277GB/s vs 83.20GB/s)
13696 additional rendering cores
Score
Benchmark
FP32 (float)
GRID M40
793
Radeon Instinct MI300
+5936%
47870
Graphics Card
May 2016
Release Date
Jan 2023
GRID
Generation
Radeon Instinct
Professional
Type
Professional
PCIe 3.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
-
Base Clock
1000 MHz
-
Boost Clock
1700 MHz
1300 MHz
Memory Clock
1600 MHz
Memory
8GB
Memory Size
128GB
GDDR5
Memory Type
HBM3
128bit
Memory Bus
8192bit
83.20GB/s
Bandwidth
3277GB/s
Render Config
-
Compute Units
220
384
Shading Units
14080
32
TMUs
880
16
ROPs
0
-
Tensor Cores
880
64 KB (per SMM)
L1 Cache
16 KB (per CU)
2 MB
L2 Cache
16 MB
Theoretical Performance
16.53 GPixel/s
Pixel Rate
0 MPixel/s
33.06 GTexel/s
Texture Rate
1496 GTexel/s
-
FP16 (half)
383.0 TFLOPS
793.3 GFLOPS
FP32 (float)
47.87 TFLOPS
24.79 GFLOPS
FP64 (double)
47.87 TFLOPS
Board Design
50W
TDP
600W
250 W
Suggested PSU
1000 W
No outputs
Outputs
No outputs
-
Power Connectors
2x 8-pin
Graphics Processor
GM107
GPU Name
Aqua Vanjaram
Maxwell
Architecture
CDNA 3.0
TSMC
Foundry
TSMC
28 nm
Process Size
5 nm
1.87 billion
Transistors
153 billion
148 mm²
Die Size
1017 mm²
Graphics Features
12 (11_0)
DirectX
N/A
4.6
OpenGL
N/A
3.0
OpenCL
3.0
1.3
Vulkan
N/A
5.0
CUDA
-
5.1
Shader Model
N/A
