NVIDIA L4 vs AMD Instinct MI300X
We compared two Professional market GPUs: 24GB VRAM L4 and 192GB VRAM AMD Instinct MI300X to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
NVIDIA L4 's Advantages
Lower TDP (72W vs 750W)
AMD Instinct MI300X 's Advantages
Released 9 months late
Boost Clock has increased by 3% (2100MHz vs 2040MHz)
More VRAM (192GB vs 24GB)
Larger VRAM bandwidth (5300GB/s vs 300.1GB/s)
12032 additional rendering cores
Score
Benchmark
FP32 (float)
L4
30.29 TFLOPS
AMD Instinct MI300X
+439%
163.4 TFLOPS
Graphics Card
Mar 2023
Release Date
Dec 2023
Tesla Ada(Lxx)
Generation
Instinct
Professional
Type
Professional
PCIe 4.0 x16
Bus Interface
PCIe 5.0 x16
Clock Speeds
795 MHz
Base Clock
1000 MHz
2040 MHz
Boost Clock
2100 MHz
1563 MHz
Memory Clock
5200 MHz
Memory
24GB
Memory Size
192GB
GDDR6
Memory Type
HBM3
192bit
Memory Bus
8192bit
300.1GB/s
Bandwidth
5300GB/s
Render Config
-
Compute Units
304
60
SM Count
-
7424
Shading Units
19456
240
TMUs
880
80
ROPs
0
240
Tensor Cores
-
60
RT Cores
-
128 KB (per SM)
L1 Cache
16 KB (per CU)
48 MB
L2 Cache
16 MB
-
-
-
Theoretical Performance
163.2 GPixel/s
Pixel Rate
0 MPixel/s
489.6 GTexel/s
Texture Rate
1496 GTexel/s
30.29 TFLOPS
FP16 (half)
1300 TFLOPS
30.29 TFLOPS
FP32 (float)
163.4 TFLOPS
473.3 GFLOPS
FP64 (double)
81.7 TFLOPS
Board Design
72W
TDP
750W
250 W
Suggested PSU
1000 W
No outputs
Outputs
No outputs
None
Power Connectors
None
Graphics Processor
AD104
GPU Name
MI300
AD104-???-A1
GPU Variant
-
Ada Lovelace
Architecture
CDNA 3.0
TSMC
Foundry
TSMC
5 nm
Process Size
5 nm
35.8 billion
Transistors
146 billion
294 mm²
Die Size
1017 mm²
Graphics Features
12 Ultimate (12_2)
DirectX
N/A
4.6
OpenGL
N/A
3.0
OpenCL
3.0
1.3
Vulkan
N/A
8.9
CUDA
-
6.8
Shader Model
N/A
