Hi Joe and Luminal, thanks for the great blog. We wanted to share our MPK project (https://github.com/mirage-project/mirage ), which is a compiler and runtime for mega-kernelizing LLMs using compiler superoptimization. We also have a paper describing the approach and results here: https://arxiv.org/pdf/2512.22219. We’d be very interested to see how Luminal compares with MPK, and would be happy to exchange ideas of course.
Hi Zhihao! I’ve seen some of the work you guys have done with Mirage, it’s very great work. I’d love to talk with you some more. Would you be able to email me at joe@luminal.com?
Hi Joe and Luminal, thanks for the great blog. We wanted to share our MPK project (https://github.com/mirage-project/mirage ), which is a compiler and runtime for mega-kernelizing LLMs using compiler superoptimization. We also have a paper describing the approach and results here: https://arxiv.org/pdf/2512.22219. We’d be very interested to see how Luminal compares with MPK, and would be happy to exchange ideas of course.
Hi Zhihao! I’ve seen some of the work you guys have done with Mirage, it’s very great work. I’d love to talk with you some more. Would you be able to email me at joe@luminal.com?