Tenstorrent Blackhole p100a vs Tenstorrent Blackhole p150a
The Blackhole p100a and the Blackhole p150a deliver near-identical FP8 throughput (664 vs 664 TFLOPS dense). The Blackhole p150a also carries 4 GB more memory (32 GB vs 28 GB).
Tenstorrent Blackhole p100a
Full specs →Tenstorrent Blackhole p150a
Full specs →Specifications
Performance (TFLOPS)
FLOPS by Precision
What actually differs
Dense throughput for the Blackhole p100a against the Blackhole p150a: FP8 664 vs 664 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.
Memory is 28 GB against 32 GB, fed at 448 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 Blackhole p100a faster than the Blackhole p150a?
They are close on paper: both deliver about 664 TFLOPS of dense FP8 throughput. Memory, bandwidth, and power are the deciding differences.
Which has more memory, the Blackhole p100a or the Blackhole p150a?
The Blackhole p150a carries 32 GB of memory versus 28 GB for the Blackhole p100a. Memory bandwidth is 448 GB/s for the Blackhole p100a and 512 GB/s for the Blackhole p150a.
How much power do the Blackhole p100a and the Blackhole p150a draw?
The Blackhole p100a is rated at 300 W TDP and the Blackhole p150a at 300 W.
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.