GPU Comparison
GEFORCE
NVIDIA L40
CORE STATE
AD102
VRAM
48 GB
CLOCK SPEED
2490 MHz
TDP
300 W
BUS WIDTH
384 bit
ARCHITECTURE
Ada Lovelace
PROCESS
5 nm
LAUNCH DATE
2022
VS
GEFORCE
Tesla M4
CORE STATE
GM206
VRAM
4 GB
CLOCK SPEED
1072 MHz
TDP
50 W
BUS WIDTH
128 bit
ARCHITECTURE
Maxwell 2.0
PROCESS
28 nm
LAUNCH DATE
2015
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
DETAILED SPECIFICATIONS
SPECIFICATION
L40
Tesla M4
Core Specs
Shading Units
18,176
1,024
-94.4%
Shaders
18,176
1,024
-94.4%
TMUs
568
64
-88.7%
ROPs
192
32
-83.3%
SM Count
142
—
Clocks
Base Clock
735 MHz
872 MHz
Boost Clock
2490 MHz
1072 MHz
Memory Clock
2250 MHz
18 Gbps effective
1375 MHz
5.5 Gbps effective
Memory
Memory Size
48 GB
4 GB
VRAM (MB)
49,152
4,096
-91.7%
Memory Type
GDDR6
GDDR5
Memory Bus
384 bit
128 bit
Bandwidth
864.0 GB/s
88.00 GB/s
Cache
L1 Cache
128 KB (per SM)
48 KB (per SMM)
L2 Cache
96 MB
1024 KB
Performance
Pixel Rate
478.1 GPixel/s
34.30 GPixel/s
Texture Rate
1,414.3 GTexel/s
68.61 GTexel/s
FP32 (TFLOPS)
90.52 TFLOPS
2.195 TFLOPS
FP64 (TFLOPS)
1,414.3 GFLOPS (1:64)
68.61 GFLOPS (1:32)
FP16 (TFLOPS)
90.52 TFLOPS (1:1)
—
AI/RT
RT Cores
142
—
Tensor Cores
568
—
Power
TDP
300 W
50 W
TDP (W)
300
50
-83.3%
Suggested PSU
700 W
250 W
Power Connectors
1x 16-pin
—
Architecture
Architecture
Ada Lovelace
Maxwell 2.0
GPU Name
AD102
GM206
Generation
Server Ada
(Lxx)
Tesla Maxwell
(Mxx)
Process Size
5 nm
28 nm
Transistors
76,300 million
2,940 million
Die Size
609 mm²
228 mm²
Foundry
TSMC
TSMC
Density
125.3M / mm²
12.9M / mm²
API Support
DirectX
12 Ultimate (12_2)
12 (12_1)
OpenGL
4.6
4.6
Vulkan
1.4
1.4
OpenCL
3.0
3.0
CUDA
8.9
5.2
Shader Model
6.8
6.8
Physical
Slot Width
Dual-slot
Single-slot
Length
267 mm
10.5 inches
—
Height
111 mm
4.4 inches
—
Outputs
4x 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 Kepler
Successor
Server Hopper
Tesla Pascal