PTXBench: Benchmarking and Adapting LLMs for GPU Kernel Optimization | Hacker News - But Better

PTXBench: Benchmarking and Adapting LLMs for GPU Kernel Optimization arxiv.org

We introduce PTXBench, a benchmark for evaluating and adapting large language models (LLMs) to use architecture-specific PTX for GPU kernel optimization. PTXBench measures functional correctness, whether selected…

3 points•matt_d•about 4 hours ago•0 comments•

#ptxbench #llms #gpu #adapting #kernel #optimization #gpu kernel #kernel optimization

0 comments

No comments yet.

Related stories

Show HN: Run Linux kernel drivers in userspace on Neptune OS, NT-like OS on seL4
Hacker News · 4 points · 10 days ago
Show HN: StarSkirmish, an arena where LLMs create StarCraft Brood War botsstarskirmish.com
Hacker News · 3 points · 6 days ago
KernelBench: Can LLMs Write GPU Kernels? – Benchmark and Toolkit, Torch –> CUDAgithub.com
Hacker News · 2 points · 1 day ago
Llms.txt: agent-optimization wishful thinkingphpied.com
Hacker News · 2 points · 3 days ago
LLMs respond differently to harmful prompts when AI watermarking is usedarstechnica.com
Ars Technica · 0 points · 15 days ago
Show HN: A Simple GPU Job Schedulergithub.com
Hacker News · 2 points · 6 days ago