inference/research
draftlow · 0.60updated 2026-10-16

NVIDIA Tesla T4

nvidia turing (TU104) 2018
This record is a draft. Not yet verified. Provisional. Treat every number on this page as provisional.

Fields

id
accelerators/nvidia-tesla-t4
Vendor
nvidia
Architecture
turing (TU104)
Release year
2018
Form factors
pcie
VRAM (GB)
16
Memory type
gddr6
Memory bus (bit)
256
Memory bandwidth (GB/s)
320
Bandwidth basis
GDDR6 10 Gbps x 256-bit = 320 GB/s (NVIDIA T4 datasheet)
TDP (W)
70
Unified memory
no
Consumer part
no
notes
Low-profile 70 W datacenter inference card, 16 GB. Widely available used and as the cheapest cloud GPU.

Notes

Low-profile 70 W datacenter inference card, 16 GB. Widely available used and as the cheapest cloud GPU.

Cross-references