Chip comparison
AWS Trainium 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.
| Field | AWS Trainium 2 | Meta MTIA v2 |
|---|---|---|
| Process node | TSMC N5 | TSMC N5 |
| TDP | 500 W | 90 W |
| Memory | 96 GB HBM3 | 128 GB LPDDR5 |
| Memory bandwidth | 2,900 GB/s | 205 GB/s |
| BF16 (dense) | 667 TFLOPS | — |
| FP8 (dense) | 1,300 TFLOPS | — |
| INT8 (dense) | 1,300 TOPS | 708 TOPS |
| Form factor | AWS Neuron 2 (Trn2 UltraServer with 64 chips per unit) | PCIe (internal Meta) |
| Announced | 2024-12-03 | 2024-04-10 |
| Released | 2025-01-01 | 2024-05-01 |
Robots running AWS Trainium 2
No robots publicly running it yet.
Robots running Meta MTIA v2
No robots publicly running it yet.
Data centers with AWS Trainium 2
Data centers with Meta MTIA v2
Common questions
AWS Trainium 2 vs Meta MTIA v2: which has more memory?
Meta MTIA v2 carries more HBM (AWS Trainium 2: 96 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 Trainium 2 vs Meta MTIA v2: which is more widely used?
AWS Trainium 2: 0 robots and 3 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 Trainium 2 full page · Meta MTIA v2 full page.