CUDA Cores vs Tensor Cores Explained

Automated excerpt

CUDA cores are general-purpose processing units. Because Tensor cores are purpose-built for exactly this, they do it dramatically faster than CUDA cores doing the same job. A GPU with fewer CUDA cores but modern Tensor cores will often beat an older GPU with more CUDA cores but no (or older) Tensor cores, for training and inference workloads.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.