GPU Comparison

NVIDIA
GEFORCE

NVIDIA L4

CORE STATE AD104
VRAM 24 GB
CLOCK SPEED 2040 MHz
TDP 72 W
BUS WIDTH 192 bit
ARCHITECTURE Ada Lovelace
nm
PROCESS 5 nm
LAUNCH DATE 2023
VS
NVIDIA
GEFORCE

Quadro K620M

CORE STATE GM108S
VRAM 2 GB
CLOCK SPEED 1124 MHz
TDP 30 W
BUS WIDTH 64 bit
ARCHITECTURE Maxwell
nm
PROCESS 28 nm
LAUNCH DATE 2015

PERFORMANCE BENCHMARKS

geekbench_opencl
140,838
5,957
geekbench_vulkan
116,491
N/A

DETAILED SPECIFICATIONS

SPECIFICATION
L4
Quadro K620M
Core Specs
Shading Units
7,424
384 -94.8%
Shaders
7,424
384 -94.8%
TMUs
240
16 -93.3%
ROPs
80
8 -90.0%
SM Count
60
Clocks
Base Clock
795 MHz
1029 MHz
Boost Clock
2040 MHz
1124 MHz
Memory Clock
1563 MHz 12.5 Gbps effective
1001 MHz 2 Gbps effective
Memory
Memory Size
24 GB
2 GB
VRAM (MB)
24,576
2,048 -91.7%
Memory Type
GDDR6
DDR3
Memory Bus
192 bit
64 bit
Bandwidth
300.1 GB/s
16.02 GB/s
Cache
L1 Cache
128 KB (per SM)
64 KB (per SMM)
L2 Cache
48 MB
1024 KB
Performance
Pixel Rate
163.2 GPixel/s
8.992 GPixel/s
Texture Rate
489.6 GTexel/s
17.98 GTexel/s
FP32 (TFLOPS)
30.29 TFLOPS
863.2 GFLOPS
FP64 (TFLOPS)
473.3 GFLOPS (1:64)
26.98 GFLOPS (1:32)
FP16 (TFLOPS)
30.29 TFLOPS (1:1)
AI/RT
RT Cores
60
Tensor Cores
240
Power
TDP
72 W
30 W
TDP (W)
72
30 -58.3%
Suggested PSU
250 W
Power Connectors
None
None
Architecture
Architecture
Ada Lovelace
Maxwell
GPU Name
AD104
GM108S
Generation
Server Ada (Lxx)
Quadro Kepler-M (Kx200M)
Process Size
5 nm
28 nm
Transistors
35,800 million
1,020 million
Die Size
294 mm²
77 mm²
Foundry
TSMC
TSMC
Density
121.8M / mm²
13.2M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 (11_0)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
3.0
3.0
CUDA
8.9
5.0
Shader Model
6.8
6.7 (5.1)
Physical
Slot Width
Single-slot
MXM Module
Length
169 mm 6.7 inches
Height
56 mm 2.2 inches
Outputs
No outputs
Portable Device Dependent
Bus Interface
PCIe 4.0 x16
MXM-A (3.0)
Other
Production
Active
End-of-life
Predecessor
Server Ampere
Quadro Fermi-M
Successor
Server Hopper
Quadro Maxwell-M
View L4 Details View Quadro K620M Details