20 GB
GDDR61560 MHz
Boost Clock70 W
TDP5 nm
ProcessShare on Social Media
Share RTX 4000 SFF Ada Generation specifications with othersTechnical Specifications
Core Specifications
Architecture:
Ada Lovelace
Process Size:
5 nm
Base Clock:
720 MHz
Boost Clock:
1560 MHz
Shader Units:
6144
SMs:
48
Tensor Cores:
192
RT Cores:
48
TMUs:
192
ROPs:
64
Memory & Performance
Memory Size:
20 GB
Memory Type:
GDDR6
Memory Bandwidth:
280.0 GB/s
Bus Interface:
PCIe 4.0 x16
TDP:
70 W
Suggested PSU:
250 W
Cache & Die Information
L1 Cache:
128 KB
L2 Cache:
48.0 MB
Die Size:
294 mm²
Transistors:
35.8 billion
Transistor Density:
121.8 million/mm²
Physical Specifications
Board Length:
168 mm
Board Width:
69 mm
Power Connectors:
None
Display Outputs:
4x mini-DisplayPort 1.4a
Quick Facts
Performance
Pixel Fill Rate
99.8 GPixel/s
Texture Fill Rate
299.5 GTexel/s
FP32 Performance
19.2 TFLOPS
CUDA Toolkit Compatibility
Cuda Compute
8.9
sm_89
Architecture
Ada Lovelace
Supported CUDA Toolkits
11.8
12.0
12.1
12.2
12.3
12.4
12.5
12.6
12.7
12.8
Ada Lovelace architecture with 3rd gen RT cores and AV1 encoding