Skip to content
Permalink

Comparing changes

Choose two branches to see what’s changed or to start a new pull request. If you need to, you can also or learn more about diff comparisons.

Open a pull request

Create a new pull request by comparing changes across two branches. If you need to, you can also . Learn more about diff comparisons here.
base repository: FlashML-org/FreeToken
Failed to load repositories. Confirm that selected base ref is valid, then try again.
Loading
base: main
Choose a base ref
...
head repository: FlashML-org/FreeToken
Failed to load repositories. Confirm that selected head ref is valid, then try again.
Loading
compare: perf/default-vendored-topk-router
Choose a head ref
Checking mergeability… Don’t worry, you can still create the pull request.
  • 1 commit
  • 3 files changed
  • 1 contributor

Commits on Sep 1, 2026

  1. perf(moe): route fused_topk through the in-repo triton router 

    The triton_kernels branch is gone: fused_topk_softmax is faster at every shape
measured on H100 (2.4-9.5x vs triton_kernels.topk, 6-17x vs pure torch,
CUDA-graph GPU time), works for any k, and needs no undeclared optional package.
Slot order for power-of-2 k now follows the descending-weight convention instead
of triton_kernels' ascending expert id; downstream expert compute sums over the
k slots, so the routing decision is unchanged.
    @jason-fxz
    jason-fxz committed Sep 1, 2026
    Configuration menu
    Copy the full SHA
    d8db64f View commit details
    Browse the repository at this point in the history
Loading