AMD Radeon Instinct MI300A vs NVIDIA GeForce RTX 4070 GDDR6
We compared two Desktop platform GPUs: 192GB VRAM Radeon Instinct MI300A and 12GB VRAM GeForce RTX 4070 GDDR6 to see which GPU has better performance in key specifications, benchmark tests, power consumption, etc.
Main Differences
AMD Radeon Instinct MI300A 's Advantages
More VRAM (192GB vs 12GB)
Larger VRAM bandwidth (10.3TB/s vs 480.0GB/s)
13568 additional rendering cores
NVIDIA GeForce RTX 4070 GDDR6 's Advantages
Released 8 months late
Boost Clock has increased by 18% (2475MHz vs 2100MHz)
Lower TDP (200W vs 750W)
Score
Benchmark
FP32 (float)
Radeon Instinct MI300A
+180%
81.72 TFLOPS
GeForce RTX 4070 GDDR6
29.15 TFLOPS
Graphics Card
Dec 2023
Release Date
Aug 2024
Radeon Instinct
Generation
GeForce 40
Desktop
Type
Desktop
PCIe 5.0 x16
Bus Interface
PCIe 4.0 x16
Clock Speeds
1000 MHz
Base Clock
1920 MHz
2100 MHz
Boost Clock
2475 MHz
2525 MHz
Memory Clock
2500 MHz
Memory
192GB
Memory Size
12GB
HBM3
Memory Type
GDDR6
8192bit
Memory Bus
192bit
10.3TB/s
Bandwidth
480.0GB/s
Render Config
304
Compute Units
-
-
SM Count
46
19456
Shading Units
5888
1216
TMUs
184
0
ROPs
64
1216
Tensor Cores
184
-
RT Cores
46
16 KB (per CU)
L1 Cache
128 KB (per SM)
16 MB
L2 Cache
36 MB
256 MB
L3 Cache
-
Theoretical Performance
0 MPixel/s
Pixel Rate
158.4 GPixel/s
2554 GTexel/s
Texture Rate
455.4 GTexel/s
653.7 TFLOPS
FP16 (half)
29.15 TFLOPS
81.72 TFLOPS
FP32 (float)
29.15 TFLOPS
81.72 TFLOPS
FP64 (double)
455.4 GFLOPS
Board Design
750W
TDP
200W
1150 W
Suggested PSU
550 W
No outputs
Outputs
1x HDMI 2.1
3x DisplayPort 1.4a
None
Power Connectors
1x 16-pin
Graphics Processor
Aqua Vanjaram
GPU Name
AD104
-
GPU Variant
AD104-2XX-A1
CDNA 3.0
Architecture
Ada Lovelace
TSMC
Foundry
TSMC
5 nm
Process Size
5 nm
153 billion
Transistors
35.8 billion
1017 mm²
Die Size
294 mm²
Graphics Features
N/A
DirectX
12 Ultimate (12_2)
N/A
OpenGL
4.6
3.0
OpenCL
3.0
N/A
Vulkan
1.3
-
CUDA
8.9
N/A
Shader Model
6.8
