Chip comparison
Meta MTIA v2 vs NVIDIA DRIVE Thor
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 DRIVE Thor |
|---|---|---|
| Process node | TSMC N5 | TSMC 4NP |
| TDP | 90 W | 130 W |
| Memory | 128 GB LPDDR5 | 128 GB LPDDR5X |
| Memory bandwidth | 205 GB/s | — |
| FP8 (dense) | — | 2,000 TFLOPS |
| INT8 (dense) | 708 TOPS | 2,000 TOPS |
| Form factor | PCIe (internal Meta) | Auto-board |
| Announced | 2024-04-10 | 2022-09-20 |
| Released | 2024-05-01 | 2025-06-01 |
Robots running Meta MTIA v2
No robots publicly running it yet.
Robots running NVIDIA DRIVE Thor
Data centers with Meta MTIA v2
Data centers with NVIDIA DRIVE Thor
No data centers publicly running it yet.
Common questions
Meta MTIA v2 vs NVIDIA DRIVE Thor: which has more memory?
NVIDIA DRIVE Thor carries more HBM (Meta MTIA v2: 128 GB LPDDR5; NVIDIA DRIVE Thor: 128 GB LPDDR5X). Memory capacity sets the largest model that fits on one chip without partitioning; memory bandwidth sets inference throughput.
Meta MTIA v2 vs NVIDIA DRIVE Thor: which is more widely used?
Meta MTIA v2: 0 robots and 2 data centers publicly running it. NVIDIA DRIVE Thor: 11 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 DRIVE Thor full page.