Skip to content

Radar ·

Inside NVIDIA GPUs: matmul kernels

An article published October 1, 2025, covering GPU architecture and kernel techniques for high-performance matrix multiplication.

Published October 1, 2025, the article covers NVIDIA GPU architecture and PTX/SASS. Its scope includes warp tiling and asynchronous tensor-core pipelines applied to matrix multiplication, or matmul, kernels.

The description highlights the relationship between GPU architecture, kernel design, and matmul performance. To examine the details and verify claims, consult the original article and compare its technical terms with relevant architecture and instruction documentation; the available description gives no numerical results or specific GPU.

Original source: Por dentro das GPUs NVIDIA: anatomia de kernels matmul de alto desempenho

Curation · AgentLog

Get the essentials

Privacy, AI engineering and security in your inbox.

At most one digest per day, when there is news. Cancel at any time. We use your account's verified email.

Privacy

Looking for something else?DocumentationContact