GPU Comparison
RADEON
AMD Radeon Pro Vega 20
CORE STATE
Vega 12
VRAM
4 GB
CLOCK SPEED
1283 MHz
TDP
100 W
BUS WIDTH
1024 bit
ARCHITECTURE
GCN 5.0
PROCESS
14 nm
LAUNCH DATE
2018
VS
GEFORCE
H200 NVL
CORE STATE
GH100
VRAM
141 GB
CLOCK SPEED
1785 MHz
TDP
600 W
BUS WIDTH
6144 bit
ARCHITECTURE
Hopper
PROCESS
5 nm
LAUNCH DATE
2024
PERFORMANCE BENCHMARKS
geekbench_metal
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
Pro Vega 20
H200 NVL
Core Specs
Shading Units
1,280
16,896
+1220.0%
Shaders
1,280
16,896
+1220.0%
TMUs
80
528
+560.0%
ROPs
32
24
-25.0%
Compute Units
20
—
SM Count
—
132
Clocks
Base Clock
815 MHz
1365 MHz
Boost Clock
1283 MHz
1785 MHz
Memory Clock
740 MHz
1480 Mbps effective
1593 MHz
6.4 Gbps effective
Memory
Memory Size
4 GB
141 GB
VRAM (MB)
4,096
144,384
+3425.0%
Memory Type
HBM2
HBM3e
Memory Bus
1024 bit
6144 bit
Bandwidth
189.4 GB/s
4.89 TB/s
Cache
L1 Cache
16 KB (per CU)
256 KB (per SM)
L2 Cache
1024 KB
50 MB
Performance
Pixel Rate
41.06 GPixel/s
42.84 GPixel/s
Texture Rate
102.6 GTexel/s
942.5 GTexel/s
FP32 (TFLOPS)
3.284 TFLOPS
60.32 TFLOPS
FP64 (TFLOPS)
205.3 GFLOPS (1:16)
30.16 TFLOPS (1:2)
FP16 (TFLOPS)
6.569 TFLOPS (2:1)
241.3 TFLOPS (4:1)
AI/RT
Tensor Cores
—
528
Power
TDP
100 W
600 W
TDP (W)
100
600
+500.0%
Suggested PSU
—
1000 W
Power Connectors
—
8-pin EPS
Architecture
Architecture
GCN 5.0
Hopper
GPU Name
Vega 12
GH100
Generation
Radeon Pro Mac
(Vega Series)
Server Hopper
(Hxx)
Process Size
14 nm
5 nm
Transistors
—
80,000 million
Die Size
—
814 mm²
Foundry
GlobalFoundries
TSMC
Density
—
98.3M / mm²
API Support
DirectX
12 (12_1)
—
OpenGL
4.6
—
Vulkan
1.3
—
OpenCL
2.1
3.0
CUDA
—
9.0
Shader Model
6.0
—
Physical
Slot Width
IGP
Dual-slot
Length
—
267 mm
10.5 inches
Height
—
111 mm
4.4 inches
Outputs
Portable Device Dependent
No outputs
Bus Interface
PCIe 3.0 x16
PCIe 5.0 x16
Other
Production
End-of-life
Active
Predecessor
—
Server Ada
Successor
—
Server Blackwell