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

Imported from official source

AI Classified by Officially

Together Compiler uses AI agents to rewrite and fuse the GPU kernels in a model’s forward pass. It beats hand-written megakernels by 28% on B200 (2,118 vs 1,649 tokens/s on Llama 3.2 1B). This is how we got from one to the other.

This is an extract. The publication continues at the source.

Read the original at the source: https://www.together.ai/blog/road-to-together-compiler

Officially imported this from Together AI’s own source and shows an extract. If you work there, claiming the profile and verifying the domain lets you choose to show the full text here.

Provenance

Organization
Together AI — imported from official source
Official source
https://www.together.ai/blog/rss.xml RSS
Imported
November 16, 2026 23:30
Versions
1 recorded
Identity
https://www.together.ai/blog/road-to-together-compiler

Officially records where a publication came from, not whether it is true. Imported records are reproduced from an organization's own official source.