Chip comparison
NVIDIA DRIVE Thor vs Tesla Dojo D1
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 | NVIDIA DRIVE Thor | Tesla Dojo D1 |
|---|---|---|
| Process node | TSMC 4NP | TSMC N7 |
| Transistors (B) | — | 50 |
| Die size | — | 645 mm² |
| TDP | 130 W | 400 W |
| Memory | 128 GB LPDDR5X | — |
| BF16 (dense) | — | 362 TFLOPS |
| FP8 (dense) | 2,000 TFLOPS | 362 TFLOPS |
| INT8 (dense) | 2,000 TOPS | — |
| Form factor | Auto-board | Training tile (25 D1 chips per tile) |
| Announced | 2022-09-20 | 2021-08-19 |
| Released | 2025-06-01 | 2023-07-01 |
Robots running NVIDIA DRIVE Thor
Robots running Tesla Dojo D1
No robots publicly running it yet.
Data centers with NVIDIA DRIVE Thor
No data centers publicly running it yet.
Data centers with Tesla Dojo D1
No data centers publicly running it yet.
Common questions
NVIDIA DRIVE Thor vs Tesla Dojo D1: which is faster for training?
NVIDIA DRIVE Thor has 5.52x the dense FP8 throughput of the other (NVIDIA DRIVE Thor: 2,000 TFLOPS; Tesla Dojo D1: 362 TFLOPS). Real-world training speed also depends on memory bandwidth, interconnect, and software stack: see the spec diff for those inputs.
NVIDIA DRIVE Thor vs Tesla Dojo D1: which is more power-efficient?
Dense FP8 TFLOPS per watt: NVIDIA DRIVE Thor 15.38 (2,000 ÷ 130 W); Tesla Dojo D1 0.91 (362 ÷ 400 W). NVIDIA DRIVE Thor wins at the die level; system-level efficiency also depends on cooling and interconnect.
NVIDIA DRIVE Thor vs Tesla Dojo D1: which is more widely used?
NVIDIA DRIVE Thor: 11 robots and 0 data centers publicly running it. Tesla Dojo D1: 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 · NVIDIA DRIVE Thor full page · Tesla Dojo D1 full page.