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
GPU
Arc A310
CORE STATE
DG2-128
VRAM
4 GB
CLOCK SPEED
1750 MHz
TDP
30 W
BUS WIDTH
64 bit
ARCHITECTURE
Xe-HPG
PROCESS
6 nm
LAUNCH DATE
2022
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
passmark_directx_10
passmark_directx_11
passmark_directx_12
passmark_directx_9
passmark_g2d
passmark_g3d
passmark_gpu_compute
DETAILED SPECIFICATIONS
SPECIFICATION
Instinct MI300X
A310
Core Specs
Shading Units
19,456
768
-96.1%
Shaders
19,456
768
-96.1%
TMUs
1,216
32
-97.4%
ROPs
0
16
+∞%
Compute Units
304
—
Execution Units
—
96
Clocks
Base Clock
1000 MHz
1750 MHz
Boost Clock
2100 MHz
1750 MHz
Memory Clock
1300 MHz 5.2 Gbps effective
1937 MHz
15.5 Gbps effective
Memory
Memory Size
192 GB
4 GB
VRAM (MB)
196,608
4,096
-97.9%
Memory Type
HBM3
GDDR6
Memory Bus
8192 bit
64 bit
Bandwidth
5.32 TB/s
124.0 GB/s
Cache
L1 Cache
16 KB (per CU)
—
L2 Cache
16 MB
4 MB
L3 Cache
256 MB
—
Performance
Pixel Rate
0 MPixel/s
28.00 GPixel/s
Texture Rate
2,553.6 GTexel/s
56.00 GTexel/s
FP32 (TFLOPS)
81.72 TFLOPS
2.688 TFLOPS
FP64 (TFLOPS)
40.86 TFLOPS (1:2)
672.0 GFLOPS (1:4)
FP16 (TFLOPS)
81.72 TFLOPS (1:1)
5.376 TFLOPS (2:1)
AI/RT
RT Cores
—
6
XMX Cores
—
96
Matrix Cores
1,216
—
Power
TDP
750 W
30 W
TDP (W)
750
30
-96.0%
Suggested PSU
1150 W
200 W
Power Connectors
None
None
Architecture
Architecture
CDNA 3.0
Xe-HPG
GPU Name
Aqua Vanjaram
DG2-128
Generation
Instinct (MIx)
Alchemist
(Arc 3)
Process Size
5 nm
6 nm
Transistors
153,000 million
7,200 million
Die Size
1017 mm²
157 mm²
Foundry
TSMC
TSMC
Density
150.4M / mm²
45.9M / mm²
AMD MCM
MCM
2
—
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
Shader Model
—
6.6
Physical
Slot Width
OAM Module
Single-slot
Outputs
No outputs
4x mini-DisplayPort 2.0
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x8
Other
Production
—
End-of-life
Predecessor
Radeon Instinct
Xe Graphics
Successor
—
Battlemage