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
L4
CORE STATE
AD104
VRAM
24 GB
CLOCK SPEED
2040 MHz
TDP
72 W
BUS WIDTH
192 bit
ARCHITECTURE
Ada Lovelace
PROCESS
5 nm
LAUNCH DATE
2023
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
Instinct MI300X
L4
Core Specs
Shading Units
19,456
7,424
-61.8%
Shaders
19,456
7,424
-61.8%
TMUs
1,216
240
-80.3%
ROPs
0
80
+∞%
Compute Units
304
—
SM Count
—
60
Clocks
Base Clock
1000 MHz
795 MHz
Boost Clock
2100 MHz
2040 MHz
Memory Clock
1300 MHz 5.2 Gbps effective
1563 MHz
12.5 Gbps effective
Memory
Memory Size
192 GB
24 GB
VRAM (MB)
196,608
24,576
-87.5%
Memory Type
HBM3
GDDR6
Memory Bus
8192 bit
192 bit
Bandwidth
5.32 TB/s
300.1 GB/s
Cache
L1 Cache
16 KB (per CU)
128 KB (per SM)
L2 Cache
16 MB
48 MB
L3 Cache
256 MB
—
Performance
Pixel Rate
0 MPixel/s
163.2 GPixel/s
Texture Rate
2,553.6 GTexel/s
489.6 GTexel/s
FP32 (TFLOPS)
81.72 TFLOPS
30.29 TFLOPS
FP64 (TFLOPS)
40.86 TFLOPS (1:2)
473.3 GFLOPS (1:64)
FP16 (TFLOPS)
81.72 TFLOPS (1:1)
30.29 TFLOPS (1:1)
AI/RT
RT Cores
—
60
Tensor Cores
—
240
Matrix Cores
1,216
—
Power
TDP
750 W
72 W
TDP (W)
750
72
-90.4%
Suggested PSU
1150 W
250 W
Power Connectors
None
None
Architecture
Architecture
CDNA 3.0
Ada Lovelace
GPU Name
Aqua Vanjaram
AD104
Generation
Instinct (MIx)
Server Ada
(Lxx)
Process Size
5 nm
5 nm
Transistors
153,000 million
35,800 million
Die Size
1017 mm²
294 mm²
Foundry
TSMC
TSMC
Density
150.4M / mm²
121.8M / mm²
AMD MCM
MCM
2
—
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
—
8.9
Shader Model
—
6.8
Physical
Slot Width
OAM Module
Single-slot
Length
—
169 mm
6.7 inches
Height
—
56 mm
2.2 inches
Outputs
No outputs
No outputs
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x16
Other
Production
—
Active
Predecessor
Radeon Instinct
Server Ampere
Successor
—
Server Hopper