Chip comparison
NVIDIA GB200 Grace Blackwell Superchip vs Tenstorrent Blackhole
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 GB200 Grace Blackwell Superchip | Tenstorrent Blackhole |
|---|---|---|
| Process node | TSMC 4NP | TSMC N6 |
| TDP | 2700 W | 300 W |
| Memory | 384 GB HBM3e | 32 GB GDDR6 |
| Memory bandwidth | 16,000 GB/s | 512 GB/s |
| FP16 (dense) | — | 745 TFLOPS |
| FP8 (dense) | 9,000 TFLOPS | — |
| INT8 (dense) | — | 2,980 TOPS |
| Form factor | Superchip (2xB200 + 1xGrace) | PCIe |
| Announced | 2024-03-18 | 2024-07-01 |
| Released | 2024-12-01 | 2025-01-01 |
Robots running NVIDIA GB200 Grace Blackwell Superchip
No robots publicly running it yet.
Robots running Tenstorrent Blackhole
No robots publicly running it yet.
Data centers with NVIDIA GB200 Grace Blackwell Superchip
Data centers with Tenstorrent Blackhole
No data centers publicly running it yet.
Common questions
NVIDIA GB200 Grace Blackwell Superchip vs Tenstorrent Blackhole: which is faster for training?
NVIDIA GB200 Grace Blackwell Superchip has 12.08x the dense FP16/BF16 throughput of the other (NVIDIA GB200 Grace Blackwell Superchip: 9,000 TFLOPS; Tenstorrent Blackhole: 745 TFLOPS). Real-world training speed also depends on memory bandwidth, interconnect, and software stack: see the spec diff for those inputs.
NVIDIA GB200 Grace Blackwell Superchip vs Tenstorrent Blackhole: which has more memory?
NVIDIA GB200 Grace Blackwell Superchip carries more HBM (NVIDIA GB200 Grace Blackwell Superchip: 384 GB HBM3e; Tenstorrent Blackhole: 32 GB GDDR6). Memory capacity sets the largest model that fits on one chip without partitioning; memory bandwidth sets inference throughput.
NVIDIA GB200 Grace Blackwell Superchip vs Tenstorrent Blackhole: which is more power-efficient?
Dense FP16/BF16 TFLOPS per watt: NVIDIA GB200 Grace Blackwell Superchip 3.33 (9,000 ÷ 2700 W); Tenstorrent Blackhole 2.48 (745 ÷ 300 W). NVIDIA GB200 Grace Blackwell Superchip wins at the die level; system-level efficiency also depends on cooling and interconnect.
NVIDIA GB200 Grace Blackwell Superchip vs Tenstorrent Blackhole: which is more widely used?
NVIDIA GB200 Grace Blackwell Superchip: 0 robots and 11 data centers publicly running it. Tenstorrent Blackhole: 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 GB200 Grace Blackwell Superchip full page · Tenstorrent Blackhole full page.