DEPLOYDatabase

Chip comparison

NVIDIA A100 Tensor Core GPU vs Tesla HW4 (FSD Computer 2)

Side-by-side specs

Straight from each vendor's datasheet. Winner in bold: higher throughput, memory, or transistor count; lower power draw or price. Sparse (2:4) throughput roughly doubles the FP8/INT8 numbers shown.

FieldNVIDIA A100 Tensor Core GPUTesla HW4 (FSD Computer 2)
Process nodeTSMC N7Samsung 7LPP
Transistors (B)54
Die size826 mm²
TDP400 W90 W
Memory80 GB HBM2e16 GB GDDR6
Memory bandwidth2,039 GB/s
FP16 (dense)312 TFLOPS
BF16 (dense)312 TFLOPS
INT8 (dense)624 TOPS300 TOPS
Launch price (list)$10,000
Form factorSXM4Auto-board (dual-SoC)
Announced2020-05-142023-01-01
Released2020-05-142023-02-01

Robots running NVIDIA A100 Tensor Core GPU

No robots publicly running it yet.

Robots running Tesla HW4 (FSD Computer 2)

Data centers with NVIDIA A100 Tensor Core GPU

No data centers publicly running it yet.

Data centers with Tesla HW4 (FSD Computer 2)

No data centers publicly running it yet.

Common questions

NVIDIA A100 Tensor Core GPU vs Tesla HW4 (FSD Computer 2): which has more memory?

NVIDIA A100 Tensor Core GPU carries more HBM (NVIDIA A100 Tensor Core GPU: 80 GB HBM2e; Tesla HW4 (FSD Computer 2): 16 GB GDDR6). Memory capacity sets the largest model that fits on one chip without partitioning; memory bandwidth sets inference throughput.

NVIDIA A100 Tensor Core GPU vs Tesla HW4 (FSD Computer 2): which is more widely used?

NVIDIA A100 Tensor Core GPU: 0 robots and 0 data centers publicly running it. Tesla HW4 (FSD Computer 2): 3 robots and 0 data centers. Empty here means no operator has said publicly, not zero adoption in the market.

See also: every chip comparison · NVIDIA A100 Tensor Core GPU full page · Tesla HW4 (FSD Computer 2) full page.