GPU Comparison
GEFORCE
NVIDIA A2
CORE STATE
GA107
VRAM
16 GB
CLOCK SPEED
1770 MHz
TDP
60 W
BUS WIDTH
128 bit
ARCHITECTURE
Ampere
PROCESS
8 nm
LAUNCH DATE
2021
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
A2
L4
Core Specs
Shading Units
1,280
7,424
+480.0%
Shaders
1,280
7,424
+480.0%
TMUs
40
240
+500.0%
ROPs
32
80
+150.0%
SM Count
10
60
+500.0%
Clocks
Base Clock
1440 MHz
795 MHz
Boost Clock
1770 MHz
2040 MHz
Memory Clock
1563 MHz
12.5 Gbps effective
1563 MHz
12.5 Gbps effective
Memory
Memory Size
16 GB
24 GB
VRAM (MB)
16,384
24,576
+50.0%
Memory Type
GDDR6
GDDR6
Memory Bus
128 bit
192 bit
Bandwidth
200.1 GB/s
300.1 GB/s
Cache
L1 Cache
128 KB (per SM)
128 KB (per SM)
L2 Cache
2 MB
48 MB
Performance
Pixel Rate
56.64 GPixel/s
163.2 GPixel/s
Texture Rate
70.80 GTexel/s
489.6 GTexel/s
FP32 (TFLOPS)
4.531 TFLOPS
30.29 TFLOPS
FP64 (TFLOPS)
70.80 GFLOPS (1:64)
473.3 GFLOPS (1:64)
FP16 (TFLOPS)
4.531 TFLOPS (1:1)
30.29 TFLOPS (1:1)
AI/RT
RT Cores
10
60
+500.0%
Tensor Cores
40
240
+500.0%
Power
TDP
60 W
72 W
TDP (W)
60
72
+20.0%
Suggested PSU
250 W
250 W
Power Connectors
None
None
Architecture
Architecture
Ampere
Ada Lovelace
GPU Name
GA107
AD104
Generation
Workstation Ampere
(Ax000)
Server Ada
(Lxx)
Process Size
8 nm
5 nm
Transistors
8,700 million
35,800 million
Die Size
200 mm²
294 mm²
Foundry
Samsung
TSMC
Density
43.5M / mm²
121.8M / 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
8.6
8.9
Shader Model
6.8
6.8
Physical
Slot Width
Single-slot
Single-slot
Length
—
169 mm
6.7 inches
Height
—
56 mm
2.2 inches
Outputs
No outputs
No outputs
Bus Interface
PCIe 4.0 x8
PCIe 4.0 x16
Other
Production
End-of-life
Active
Predecessor
Quadro Turing
Server Ampere
Successor
Workstation Ada
Server Hopper