GPU Comparison
GEFORCE
NVIDIA L40S
CORE STATE
AD102
VRAM
48 GB
CLOCK SPEED
2520 MHz
TDP
300 W
BUS WIDTH
384 bit
ARCHITECTURE
Ada Lovelace
PROCESS
5 nm
LAUNCH DATE
2022
VS
GEFORCE
Tesla T4
CORE STATE
TU104
VRAM
16 GB
CLOCK SPEED
1590 MHz
TDP
70 W
BUS WIDTH
256 bit
ARCHITECTURE
Turing
PROCESS
12 nm
LAUNCH DATE
2018
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
L40S
Tesla T4
Core Specs
Shading Units
18,176
2,560
-85.9%
Shaders
18,176
2,560
-85.9%
TMUs
568
160
-71.8%
ROPs
192
64
-66.7%
SM Count
142
40
-71.8%
Clocks
Base Clock
1110 MHz
585 MHz
Boost Clock
2520 MHz
1590 MHz
Memory Clock
2250 MHz
18 Gbps effective
1250 MHz
10 Gbps effective
Memory
Memory Size
48 GB
16 GB
VRAM (MB)
49,152
16,384
-66.7%
Memory Type
GDDR6
GDDR6
Memory Bus
384 bit
256 bit
Bandwidth
864.0 GB/s
320.0 GB/s
Cache
L1 Cache
128 KB (per SM)
64 KB (per SM)
L2 Cache
48 MB
4 MB
Performance
Pixel Rate
483.8 GPixel/s
101.8 GPixel/s
Texture Rate
1,431.4 GTexel/s
254.4 GTexel/s
FP32 (TFLOPS)
91.61 TFLOPS
8.141 TFLOPS
FP64 (TFLOPS)
1,431.4 GFLOPS (1:64)
254.4 GFLOPS (1:32)
FP16 (TFLOPS)
91.61 TFLOPS (1:1)
65.13 TFLOPS (8:1)
AI/RT
RT Cores
142
40
-71.8%
Tensor Cores
568
320
-43.7%
Power
TDP
300 W
70 W
TDP (W)
300
70
-76.7%
Suggested PSU
700 W
250 W
Power Connectors
1x 16-pin
None
Architecture
Architecture
Ada Lovelace
Turing
GPU Name
AD102
TU104
Generation
Server Ada
(Lxx)
Tesla Turing
(Txx)
Process Size
5 nm
12 nm
Transistors
76,300 million
13,600 million
Die Size
609 mm²
545 mm²
Foundry
TSMC
TSMC
Density
125.3M / mm²
25.0M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 Ultimate (12_2)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
3.0
3.0
CUDA
8.9
7.5
Shader Model
6.8
6.8
Physical
Slot Width
Dual-slot
Single-slot
Length
267 mm
10.5 inches
168 mm
6.6 inches
Height
111 mm
4.4 inches
—
Outputs
1x HDMI 2.13x DisplayPort 1.4a
No outputs
Bus Interface
PCIe 4.0 x16
PCIe 3.0 x16
Other
Production
End-of-life
End-of-life
Predecessor
Server Ampere
Tesla Volta
Successor
Server Hopper
Server Ampere