GPU Comparison
RADEON
AMD Radeon Instinct MI25
CORE STATE
Vega 10
VRAM
16 GB
CLOCK SPEED
1500 MHz
TDP
300 W
BUS WIDTH
2048 bit
ARCHITECTURE
GCN 5.0
PROCESS
14 nm
LAUNCH DATE
2017
VS
GEFORCE
L20
CORE STATE
AD102
VRAM
48 GB
CLOCK SPEED
2520 MHz
TDP
275 W
BUS WIDTH
384 bit
ARCHITECTURE
Ada Lovelace
PROCESS
5 nm
LAUNCH DATE
2023
PERFORMANCE BENCHMARKS
geekbench_opencl
DETAILED SPECIFICATIONS
SPECIFICATION
Instinct MI25
L20
Core Specs
Shading Units
4,096
11,776
+187.5%
Shaders
4,096
11,776
+187.5%
TMUs
256
368
+43.8%
ROPs
64
128
+100.0%
Compute Units
64
—
SM Count
—
92
Clocks
Base Clock
1400 MHz
1440 MHz
Boost Clock
1500 MHz
2520 MHz
Memory Clock
852 MHz
1704 Mbps effective
2250 MHz
18 Gbps effective
Memory
Memory Size
16 GB
48 GB
VRAM (MB)
16,384
49,152
+200.0%
Memory Type
HBM2
GDDR6
Memory Bus
2048 bit
384 bit
Bandwidth
436.2 GB/s
864.0 GB/s
Cache
L1 Cache
16 KB (per CU)
128 KB (per SM)
L2 Cache
4 MB
96 MB
Performance
Pixel Rate
96.00 GPixel/s
322.6 GPixel/s
Texture Rate
384.0 GTexel/s
927.4 GTexel/s
FP32 (TFLOPS)
12.29 TFLOPS
59.35 TFLOPS
FP64 (TFLOPS)
768.0 GFLOPS (1:16)
927.4 GFLOPS (1:64)
FP16 (TFLOPS)
24.58 TFLOPS (2:1)
59.35 TFLOPS (1:1)
AI/RT
RT Cores
—
92
Tensor Cores
—
368
Power
TDP
300 W
275 W
TDP (W)
300
275
-8.3%
Suggested PSU
700 W
600 W
Power Connectors
2x 8-pin
1x 16-pin
Architecture
Architecture
GCN 5.0
Ada Lovelace
GPU Name
Vega 10
AD102
Generation
Radeon Instinct
(MIx)
Server Ada
(Lxx)
Process Size
14 nm
5 nm
Transistors
12,500 million
76,300 million
Die Size
495 mm²
609 mm²
Foundry
GlobalFoundries
TSMC
Density
25.3M / mm²
125.3M / mm²
API Support
DirectX
12 (12_1)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.3
1.4
OpenCL
2.1
3.0
CUDA
—
8.9
Shader Model
6.7
6.8
Physical
Slot Width
Dual-slot
Dual-slot
Length
267 mm
10.5 inches
267 mm
10.5 inches
Height
111 mm
4.4 inches
111 mm
4.4 inches
Outputs
No outputs
4x DisplayPort 1.4a
Bus Interface
PCIe 3.0 x16
PCIe 4.0 x16
Other
Production
End-of-life
Active
Predecessor
FirePro Data Center
Server Ampere
Successor
—
Server Hopper