AutoMegaKernel
An agent harness that compiles a model into one provably-correct, self-retargeting CUDA megakernel and self-tunes it past cuBLAS at batch-1 LLM decode, paper: https://arxiv.org/abs/2606.09682
RightNow-AI/AutoMegaKernel
View on GitHub ↗An agent harness that compiles a model into one provably-correct, self-retargeting CUDA megakernel and self-tunes it past cuBLAS at batch-1 LLM decode, paper: https://arxiv.org/abs/2606.09682
AutoMegaKernel grew +9.8% in the last 30 days, ranks 207th of 303 in Inference Optimization, last commit 37 days ago.
Star trajectory
109 days tracked+9.8% in 30d, above the Inference Optimization median of +1.5%.
Last commit 37 days ago, staler than the median of 15d.
How it compares in Inference Optimization
Ranked 207th of 303 by stars.
- Language
- Python
- License
- MIT License
- Contributors
- 1
- Forks
- 15
- Watchers
- 3
- Open issues
- 2
- Releases
- 0
- Created
- 2026-06-08
- Stars this week
- +5
You've seen where AutoMegaKernel stands today.
Get the moves before everyone else: the 5 Inference Optimization repos climbing fastest, every Tuesday.
Free weekly digest. Double opt-in, unsubscribe anytime.