Chip comparison
Meta MTIA v2 vs NVIDIA B100 (Blackwell)
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 | Meta MTIA v2 | NVIDIA B100 (Blackwell) |
|---|---|---|
| Process node | TSMC N5 | TSMC 4NP |
| TDP | 90 W | 700 W |
| Memory | 128 GB LPDDR5 | 192 GB HBM3e |
| Memory bandwidth | 205 GB/s | 8,000 GB/s |
| FP16 (dense) | — | 1,750 TFLOPS |
| BF16 (dense) | — | 1,750 TFLOPS |
| FP8 (dense) | — | 3,500 TFLOPS |
| INT8 (dense) | 708 TOPS | — |
| Form factor | PCIe (internal Meta) | SXM6 (700 W envelope) |
| Announced | 2024-04-10 | 2024-03-18 |
| Released | 2024-05-01 | 2024-11-01 |
Robots running Meta MTIA v2
No robots publicly running it yet.
Robots running NVIDIA B100 (Blackwell)
No robots publicly running it yet.
Data centers with Meta MTIA v2
Data centers with NVIDIA B100 (Blackwell)
No data centers publicly running it yet.
Common questions
Meta MTIA v2 vs NVIDIA B100 (Blackwell): which has more memory?
NVIDIA B100 (Blackwell) carries more HBM (Meta MTIA v2: 128 GB LPDDR5; NVIDIA B100 (Blackwell): 192 GB HBM3e). Memory capacity sets the largest model that fits on one chip without partitioning; memory bandwidth sets inference throughput.
Meta MTIA v2 vs NVIDIA B100 (Blackwell): which is more widely used?
Meta MTIA v2: 0 robots and 2 data centers publicly running it. NVIDIA B100 (Blackwell): 0 robots and 0 data centers. Empty here means no operator has said publicly, not zero adoption in the market.
See also: every chip comparison · Meta MTIA v2 full page · NVIDIA B100 (Blackwell) full page.