Skip to content
HardwareLadder
AI GPU ladder

NVIDIAAI GPU · PCIe (low profile)

T4

NVIDIA T4 is a Turing data-center accelerator with 16 GB GDDR6 and 0.32 TB/s of memory bandwidth.

Ladder rank
#25
AI index (H100 SXM = 100)
7.9
Series
Volta / Turing

Specifications

Dense throughput, no sparsity
Architecture Turing
Form factor PCIe (low profile)
Process TSMC 12 nm
FP16 / BF16 tensor (dense) 65 TFLOPS FP16 only (no BF16 support)
FP8 tensor (dense) —
FP4 tensor (dense) —
Memory 16 GB GDDR6
Memory bandwidth 0.32 TB/s
Power 70 W
Launch Sep 2018

AI index

Tensor × bandwidth

100 × √((65 ÷ 989.4) × (0.32 ÷ 3.35)) = 7.9

The AI index is the geometric mean of dense 16-bit tensor throughput and memory bandwidth, each relative to the H100 SXM. It is a spec-sheet comparison, not a measured benchmark.

Sources

2 references