GPU Comparison
RADEON
AMD Instinct MI300X
CORE STATE
Aqua Vanjaram
VRAM
192 GB
CLOCK SPEED
2100 MHz
TDP
750 W
BUS WIDTH
8192 bit
ARCHITECTURE
CDNA 3.0
PROCESS
5 nm
LAUNCH DATE
2023
VS
GEFORCE
CMP 40HX
CORE STATE
TU106
VRAM
8 GB
CLOCK SPEED
1650 MHz
TDP
185 W
BUS WIDTH
256 bit
ARCHITECTURE
Turing
PROCESS
12 nm
LAUNCH DATE
2021
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
Instinct MI300X
CMP 40HX
Core Specs
Shading Units
19,456
2,304
-88.2%
Shaders
19,456
2,304
-88.2%
TMUs
1,216
144
-88.2%
ROPs
0
64
+∞%
Compute Units
304
—
SM Count
—
36
Clocks
Base Clock
1000 MHz
1470 MHz
Boost Clock
2100 MHz
1650 MHz
Memory Clock
1300 MHz 5.2 Gbps effective
1750 MHz
14 Gbps effective
Memory
Memory Size
192 GB
8 GB
VRAM (MB)
196,608
8,192
-95.8%
Memory Type
HBM3
GDDR6
Memory Bus
8192 bit
256 bit
Bandwidth
5.32 TB/s
448.0 GB/s
Cache
L1 Cache
16 KB (per CU)
64 KB (per SM)
L2 Cache
16 MB
4 MB
L3 Cache
256 MB
—
Performance
Pixel Rate
0 MPixel/s
105.6 GPixel/s
Texture Rate
2,553.6 GTexel/s
237.6 GTexel/s
FP32 (TFLOPS)
81.72 TFLOPS
7.603 TFLOPS
FP64 (TFLOPS)
40.86 TFLOPS (1:2)
237.6 GFLOPS (1:32)
FP16 (TFLOPS)
81.72 TFLOPS (1:1)
15.21 TFLOPS (2:1)
AI/RT
RT Cores
—
36
Tensor Cores
—
288
Matrix Cores
1,216
—
Power
TDP
750 W
185 W
TDP (W)
750
185
-75.3%
Suggested PSU
1150 W
450 W
Power Connectors
None
1x 8-pin
Architecture
Architecture
CDNA 3.0
Turing
GPU Name
Aqua Vanjaram
TU106
Generation
Instinct (MIx)
Mining GPUs
Process Size
5 nm
12 nm
Transistors
153,000 million
10,800 million
Die Size
1017 mm²
445 mm²
Foundry
TSMC
TSMC
Density
150.4M / mm²
24.3M / mm²
AMD MCM
MCM
2
—
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
—
7.5
Shader Model
—
6.8
Physical
Slot Width
OAM Module
Dual-slot
Length
—
229 mm
9 inches
Height
—
111 mm
4.4 inches
Outputs
No outputs
No outputs
Bus Interface
PCIe 5.0 x16
PCIe 1.0 x4
Other
Launch Price
—
699 USD
Production
—
End-of-life
Predecessor
Radeon Instinct
—