Graphcore GC200 vs Graphcore Bow IPU
The Bow IPU delivers 40% more FP16 throughput than the GC200 (350 vs 250 TFLOPS dense).
Graphcore GC200
Full specs →Graphcore Bow IPU
Full specs →Specifications
Performance (TFLOPS)
FLOPS by Precision
What actually differs
The Bow IPU is the newer part: Colossus MK2, launched in 2022, against the GC200's Colossus MK2 from 2020. Newer architectures typically add lower-precision formats and better throughput per watt, so check the precision rows your workload actually uses.
Dense throughput for the GC200 against the Bow IPU: FP32 63 vs 87 TFLOPS, FP16 250 vs 350 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.
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 GC200 faster than the Bow IPU?
At FP16 precision the Bow IPU reaches 350 TFLOPS dense against 250 TFLOPS for the GC200. The performance table on this page lists every published precision for both GPUs.
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.