- 70 W power consumption for energy-efficient operation
- 320 GBps bandwidth with GDDR6 SDRAM for fast data processing
- Fanless design for a quiet and dust-free environment
- Supports DirectCompute, OpenACC, and OpenCL APIs for versatile computing
- 2560 CUDA cores and NVIDIA Turing architecture for significant computing performance
The NVIDIA Tesla T4 GPU is a single-slot, low-profile PCIe 3.0 universal deep learning accelerator. It provides multi-precision performance to accelerate deep learning and machine learning training and inference, video transcoding and virtual desktops.