GPU Comparison
GEFORCE
NVIDIA GB10
CORE STATE
GB20B
VRAM
128 GB
CLOCK SPEED
2418 MHz
TDP
140 W
BUS WIDTH
256 bit
ARCHITECTURE
Blackwell 2.0
PROCESS
5 nm
LAUNCH DATE
2025
VS
GEFORCE
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
PERFORMANCE BENCHMARKS
geekbench_opencl
geekbench_vulkan
Analysis: NVIDIA GB10 vs NVIDIA L40
Looking at the test data for NVIDIA's NVIDIA L40 graphics and NVIDIA GB10 (NVIDIA), a pattern of competitive throughput develops. None of the graphics card wins steadily in the benchmarks. The back-and-forth of advantages implies core equivalence. Buyers confront a genuine choice with these options. The two cards achieve solid graphics results.
In the end, builders could pick based on price rather than absolute performance exclusively. Each NVIDIA's NVIDIA L40 graphics and NVIDIA GB10 (NVIDIA) provide solid graphics results for their given price points. The selection rests on personal requirements. Each choice is reasonable. You won't go wrong between them.
DETAILED SPECIFICATIONS
SPECIFICATION
GB10
L40
Core Specs
Shading Units
6,144
18,176
+195.8%
Shaders
6,144
18,176
+195.8%
TMUs
384
568
+47.9%
ROPs
48
192
+300.0%
SM Count
48
142
+195.8%
Clocks
Base Clock
1665 MHz
735 MHz
Boost Clock
2418 MHz
2490 MHz
Memory Clock
1067 MHz
8.5 Gbps effective
2250 MHz
18 Gbps effective
Memory
Memory Size
128 GB
48 GB
VRAM (MB)
131,072
49,152
-62.5%
Memory Type
LPDDR5X
GDDR6
Memory Bus
256 bit
384 bit
Bandwidth
273.2 GB/s
864.0 GB/s
Cache
L1 Cache
256 KB (per SM)
128 KB (per SM)
L2 Cache
50 MB
96 MB
Performance
Pixel Rate
116.1 GPixel/s
478.1 GPixel/s
Texture Rate
928.5 GTexel/s
1,414.3 GTexel/s
FP32 (TFLOPS)
29.71 TFLOPS
90.52 TFLOPS
FP64 (TFLOPS)
464.3 GFLOPS (1:64)
1,414.3 GFLOPS (1:64)
FP16 (TFLOPS)
29.71 TFLOPS (1:1)
90.52 TFLOPS (1:1)
AI/RT
RT Cores
48
142
+195.8%
Tensor Cores
384
568
+47.9%
Power
TDP
140 W
300 W
TDP (W)
140
300
+114.3%
Suggested PSU
300 W
700 W
Power Connectors
None
1x 16-pin
Architecture
Architecture
Blackwell 2.0
Ada Lovelace
GPU Name
GB20B
AD102
Generation
Server Blackwell
(Bxx)
Server Ada
(Lxx)
Process Size
5 nm
5 nm
Transistors
—
76,300 million
Die Size
—
609 mm²
Foundry
TSMC
TSMC
Density
—
125.3M / mm²
API Support
DirectX
—
12 Ultimate (12_2)
OpenGL
—
4.6
Vulkan
—
1.4
OpenCL
3.0
3.0
CUDA
12.1
8.9
Shader Model
—
6.8
Physical
Slot Width
IGP
Dual-slot
Length
150 mm
5.9 inches
267 mm
10.5 inches
Height
51 mm
2 inches
111 mm
4.4 inches
Outputs
1x HDMI
4x DisplayPort 1.4a
Bus Interface
PCIe 5.0 x16
PCIe 4.0 x16
Other
Launch Price
3,999 USD
—
Production
Active
End-of-life
Predecessor
Server Hopper
Server Ampere
Successor
Server Rubin
Server Hopper