FlashQLA
Alibaba CloudFlashQLA is a linear attention kernel library from the Qwen team, built on TileLang, applying operator fusion to the forward and backward passes of Gated DeltaNet (GDN) chunked prefill. It serves as a backend for the flash-linear-attention library's GDN operators through the standard FLA API.
Openness
5 high confidence- 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/QwenLM/FlashQLA recorded 2026-09-02
Repo metadata for flashqla's canonical repository - private false, archived false, fork false.
- https://github.com/QwenLM/FlashQLA/blob/main/LICENSE recorded 2026-10-07
LICENSE body is the verbatim MIT text under an Alibaba/Qwen copyright line.
- https://github.com/QwenLM/FlashQLA/blob/main/README.md recorded 2026-10-07
README describes FlashQLA as "a high-performance linear attention kernel library built on" TileLang, which "now serves as a backend for" flash-linear-attention's Gated DeltaNet; it names no paid tier, enterprise edition or license-gated build beside the published source.
Adoption
1 high confidencePyPI downloads of `flash-qla` are the adoption signal here; its README documents `pip install flash-qla` as the install path, and the package description links back to the QwenLM/FlashQLA repository.
- https://pypi.org/pypi/flash-qla/json recorded 2026-09-02
PyPI metadata for flash-qla 0.1.2 - summary 'FlashQLA: Fused TileLang kernels for Linear Attention', description links https://github.com/QwenLM/FlashQLA, requires tilelang.
- https://pypistats.org/api/packages/flash-qla/recent recorded 2026-09-02
last_month downloads = 2,326 for flash-qla
Capability
2 medium confidenceFlashQLA provides forward and backward kernels for one linear-attention variant's chunked-prefill path, Gated DeltaNet, built as a drop-in backend for the flash-linear-attention library's API. That single-variant, single-integration scope is well short of a production engine such as TensorRT.
- https://github.com/QwenLM/FlashQLA/blob/main/README.md recorded 2026-10-07
README describes FlashQLA as "a high-performance linear attention kernel library built on" TileLang, which "now serves as a backend for" flash-linear-attention's Gated DeltaNet; it names no paid tier, enterprise edition or license-gated build beside the published source.
Verified 2026-09-24