FlashKDA

license

MIT(OSI)

source

public

core features withheld

no

MIT license body confirmed. The repository is public and unarchived and builds the whole product, and the README describes no paid tier, enterprise edition or license-gated build beside it, so source is public and the core ungated.

• https://api.github.com/repos/MoonshotAI/FlashKDA recorded 2026-09-02

  Repo metadata for flashkda's canonical repository - private false, archived false, fork false.

• https://github.com/MoonshotAI/FlashKDA/blob/master/LICENSE recorded 2026-10-07

  LICENSE body is the verbatim MIT text under a Moonshot AI copyright line.

• https://github.com/MoonshotAI/FlashKDA/blob/master/README.md recorded 2026-10-07

  README describes FlashKDA as high-performance Kimi Delta Attention kernels built on CUTLASS, auto-dispatched from flash-linear-attention's chunk_kda, with no paid tier, enterprise edition or license-gated build beside the published source.

FlashKDA ships no registry package and installs from source, so GitHub stars are the only available adoption signal.

• https://api.github.com/repos/MoonshotAI/FlashKDA recorded 2026-09-02

  Repo metadata - stargazers_count = 1,243 - for MoonshotAI/FlashKDA.

FlashKDA provides forward and backward CUTLASS kernels for one linear-attention variant, Kimi Delta Attention, purpose-built for Moonshot AI's Kimi model family rather than a broad attention technique. That single-model, single-variant scope is well short of a production engine such as TensorRT.

• https://github.com/MoonshotAI/FlashKDA/blob/master/README.md recorded 2026-10-07

  README describes FlashKDA as high-performance Kimi Delta Attention kernels built on CUTLASS, auto-dispatched from flash-linear-attention's chunk_kda, with no paid tier, enterprise edition or license-gated build beside the published source.