Skip to content
AI Primer

KernelBench

Can LLMs Write Efficient GPU Kernels?

Open-source benchmark and evaluation environment for testing large language models' ability to generate correct and efficient GPU kernels from PyTorch programs, with tasks spanning operators, fusion patterns, model architectures, and Hugging Face architectures.

Screenshot of KernelBench website

Recent stories

0 linked stories
No linked stories yet.
AI PrimerAI Primer

Your daily guide to AI tools, workflows, and creative inspiration.

© 2026 AI Primer. All rights reserved.