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
RADEON
Radeon Pro 450
CORE STATE
Baffin
VRAM
2 GB
CLOCK SPEED
—
TDP
35 W
BUS WIDTH
128 bit
ARCHITECTURE
GCN 4.0
PROCESS
14 nm
LAUNCH DATE
2016
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_metal
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
Instinct MI300X
Pro 450
Core Specs
Shading Units
19,456
640
-96.7%
Shaders
19,456
640
-96.7%
TMUs
1,216
40
-96.7%
ROPs
0
16
+∞%
Compute Units
304
10
-96.7%
Clocks
Base Clock
1000 MHz
—
Boost Clock
2100 MHz
—
GPU Clock
—
800 MHz
Memory Clock
1300 MHz 5.2 Gbps effective
1270 MHz
5.1 Gbps effective
Memory
Memory Size
192 GB
2 GB
VRAM (MB)
196,608
2,048
-99.0%
Memory Type
HBM3
GDDR5
Memory Bus
8192 bit
128 bit
Bandwidth
5.32 TB/s
81.28 GB/s
Cache
L1 Cache
16 KB (per CU)
16 KB (per CU)
L2 Cache
16 MB
1024 KB
L3 Cache
256 MB
—
Performance
Pixel Rate
0 MPixel/s
12.80 GPixel/s
Texture Rate
2,553.6 GTexel/s
32.00 GTexel/s
FP32 (TFLOPS)
81.72 TFLOPS
1,024.0 GFLOPS
FP64 (TFLOPS)
40.86 TFLOPS (1:2)
64.00 GFLOPS (1:16)
FP16 (TFLOPS)
81.72 TFLOPS (1:1)
1,024.0 GFLOPS (1:1)
AI/RT
Matrix Cores
1,216
—
Power
TDP
750 W
35 W
TDP (W)
750
35
-95.3%
Suggested PSU
1150 W
—
Power Connectors
None
—
Architecture
Architecture
CDNA 3.0
GCN 4.0
GPU Name
Aqua Vanjaram
Baffin
Generation
Instinct (MIx)
Radeon Pro Mac
(400 Series)
Process Size
5 nm
14 nm
Transistors
153,000 million
3,000 million
Die Size
1017 mm²
123 mm²
Foundry
TSMC
GlobalFoundries
Density
150.4M / mm²
24.4M / mm²
AMD MCM
MCM
2
—
API Support
DirectX
—
12 (12_0)
OpenGL
—
4.6
Vulkan
—
1.3
OpenCL
3.0
2.1
Shader Model
—
6.7
Physical
Slot Width
OAM Module
MXM Module
Outputs
No outputs
Portable Device Dependent
Bus Interface
PCIe 5.0 x16
PCIe 3.0 x8
Other
Production
—
End-of-life
Predecessor
Radeon Instinct
—