Chip comparison
Huawei Ascend 910D 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 | Huawei Ascend 910D | Meta MTIA v2 |
|---|---|---|
| Process node | SMIC N+2 (~7nm) | TSMC N5 |
| TDP | 800 W | 90 W |
| Memory | 192 GB HBM3 | 128 GB LPDDR5 |
| Memory bandwidth | 4,000 GB/s | 205 GB/s |
| FP16 (dense) | 1,200 TFLOPS | — |
| BF16 (dense) | 1,200 TFLOPS | — |
| FP8 (dense) | 2,400 TFLOPS | — |
| INT8 (dense) | — | 708 TOPS |
| Form factor | OAM | PCIe (internal Meta) |
| Announced | 2025-04-01 | 2024-04-10 |
| Released | 2025-12-01 | 2024-05-01 |
Robots running Huawei Ascend 910D
No robots publicly running it yet.
Robots running Meta MTIA v2
No robots publicly running it yet.
Data centers with Huawei Ascend 910D
No data centers publicly running it yet.
Data centers with Meta MTIA v2
Common questions
Huawei Ascend 910D vs Meta MTIA v2: which has more memory?
Huawei Ascend 910D carries more HBM (Huawei Ascend 910D: 192 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.
Huawei Ascend 910D vs Meta MTIA v2: which is more widely used?
Huawei Ascend 910D: 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 · Huawei Ascend 910D full page · Meta MTIA v2 full page.