The story’s score and claims

Comments on No. 64 of 142 · together.ai · edition of Wednesday 18 November 2026

Last year we built megakernels. This year we deleted them: The road to TKCC

Summary

Together AI's engineering team reports that after building "megakernels" that ran an entire model forward pass in one fused GPU kernel, they found most of the gains could be recovered with simpler tools: CUDA Graphs and PDL on top of default Torch Inductor kernels recovered 98% of megakernel performance, and…

Comments

Ranked by a public rubric, not by votes: what a comment adds counts, not who wrote it. How comments are ranked · Comment rules