GPU Comparison
GEFORCE
NVIDIA 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
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
H200 NVL
L4
Core Specs
Shading Units
16,896
7,424
-56.1%
Shaders
16,896
7,424
-56.1%
TMUs
528
240
-54.5%
ROPs
24
80
+233.3%
SM Count
132
60
-54.5%
Clocks
Base Clock
1365 MHz
795 MHz
Boost Clock
1785 MHz
2040 MHz
Memory Clock
1593 MHz
6.4 Gbps effective
1563 MHz
12.5 Gbps effective
Memory
Memory Size
141 GB
24 GB
VRAM (MB)
144,384
24,576
-83.0%
Memory Type
HBM3e
GDDR6
Memory Bus
6144 bit
192 bit
Bandwidth
4.89 TB/s
300.1 GB/s
Cache
L1 Cache
256 KB (per SM)
128 KB (per SM)
L2 Cache
50 MB
48 MB
Performance
Pixel Rate
42.84 GPixel/s
163.2 GPixel/s
Texture Rate
942.5 GTexel/s
489.6 GTexel/s
FP32 (TFLOPS)
60.32 TFLOPS
30.29 TFLOPS
FP64 (TFLOPS)
30.16 TFLOPS (1:2)
473.3 GFLOPS (1:64)
FP16 (TFLOPS)
241.3 TFLOPS (4:1)
30.29 TFLOPS (1:1)
AI/RT
RT Cores
—
60
Tensor Cores
528
240
-54.5%
Power
TDP
600 W
72 W
TDP (W)
600
72
-88.0%
Suggested PSU
1000 W
250 W
Power Connectors
8-pin EPS
None
Architecture
Architecture
Hopper
Ada Lovelace
GPU Name
GH100
AD104
Generation
Server Hopper
(Hxx)
Server Ada
(Lxx)
Process Size
5 nm
5 nm
Transistors
80,000 million
35,800 million
Die Size
814 mm²
294 mm²
Foundry
TSMC
TSMC
Density
98.3M / mm²
121.8M / mm²
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
9.0
8.9
Shader Model
—
6.8
Physical
Slot Width
Dual-slot
Single-slot
Length
267 mm
10.5 inches
169 mm
6.7 inches
Height
111 mm
4.4 inches
56 mm
2.2 inches
Outputs
No outputs
No outputs
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x16
Other
Production
Active
Active
Predecessor
Server Ada
Server Ampere
Successor
Server Blackwell
Server Hopper