Tenstorrent Wormhole n300s vs Tenstorrent Blackhole p150a
The Blackhole delivers 42% more FP8 throughput than the Wormhole (664 vs 466 TFLOPS dense). The Blackhole also carries 8 GB more memory (32 GB vs 24 GB).
Tenstorrent Wormhole n300s
Full specs →Tenstorrent Blackhole p150a
Full specs →Specifications
Performance (TFLOPS)
FLOPS by Precision
What actually differs
Dense throughput for the Wormhole against the Blackhole: FP8 466 vs 664 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.
Memory is 24 GB against 32 GB, fed at 576 GB/s versus 512 GB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.
For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.
Frequently asked questions
Is the Wormhole faster than the Blackhole?
At FP8 precision the Blackhole reaches 664 TFLOPS dense against 466 TFLOPS for the Wormhole. The performance table on this page lists every published precision for both GPUs.
Which has more memory, the Wormhole or the Blackhole?
The Blackhole carries 32 GB of VRAM versus 24 GB for the Wormhole. Memory bandwidth is 576 GB/s for the Wormhole and 512 GB/s for the Blackhole.
How much power do the Wormhole and the Blackhole draw?
The Wormhole is rated at 300 W TDP and the Blackhole at 300 W.
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.