DEPLOYDatabase

Chip comparison

AWS Inferentia 2 vs Meta MTIA v2

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.

FieldAWS Inferentia 2Meta MTIA v2
Process nodeTSMC N7TSMC N5
TDP250 W90 W
Memory32 GB HBM3128 GB LPDDR5
Memory bandwidth820 GB/s205 GB/s
BF16 (dense)190 TFLOPS
FP8 (dense)380 TFLOPS
INT8 (dense)708 TOPS
Form factorInf2 EC2 instancePCIe (internal Meta)
Announced2022-12-012024-04-10
Released2023-04-012024-05-01

Robots running AWS Inferentia 2

No robots publicly running it yet.

Robots running Meta MTIA v2

No robots publicly running it yet.

Data centers with AWS Inferentia 2

No data centers publicly running it yet.

Common questions

AWS Inferentia 2 vs Meta MTIA v2: which has more memory?

Meta MTIA v2 carries more HBM (AWS Inferentia 2: 32 GB HBM3; Meta MTIA v2: 128 GB LPDDR5). Memory capacity sets the largest model that fits on one chip without partitioning; memory bandwidth sets inference throughput.

AWS Inferentia 2 vs Meta MTIA v2: which is more widely used?

AWS Inferentia 2: 0 robots and 0 data centers publicly running it. Meta MTIA v2: 0 robots and 2 data centers. Empty here means no operator has said publicly, not zero adoption in the market.

See also: every chip comparison · AWS Inferentia 2 full page · Meta MTIA v2 full page.