Repository navigation
[kimi k3] Allow Attention Gym KDA on CUDA capability 9.0+ - #4868
Merged
Merged
Conversation
KDAKernel only accepted SM100/SM103. That gate matched attn-gym 0.0.5, pinned when pytorch#4351 moved KDA onto Attention Gym. Since 0.0.7, chunk_kda runs CuTe kernels on SM100/SM103 and Triton kernels on other NVIDIA GPUs, and the pinned 0.0.13 keeps that dispatch. The remaining hardware floor is the fused bounded gate, which needs TMA (SM90+). - Relax the KDAKernel gate and the two KDA GPU test skips to >= (9, 0). - Document the hardware requirement in the Kimi K3 README.
JavaZeroo
requested review from
fegin,
tianyu-l,
wconstab and
wwwjn
as code owners
September 24, 2026 17:30
meta-cla
Bot
added
the
CLA Signed
|
|
drisspg
approved these changes
Sep 24, 2026
tianyu-l
approved these changes
Sep 24, 2026
Contributor
Author
yes, I just tested on rtx5080 only. |
Contributor
|
Yes, A100 and RTX5060Ti also works actually, I have ran on all these, as long as the eager impl in AttnGym is available |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Currently,
KDAKernelonly accept SM100/SM103, so Kimi K3 and its KDA tests only run on B200-class GPUs. This gate is stale:attn-gym==0.0.5pin, when the KDA kernels were Blackwell-only.chunk_kdauses CuTe kernels on SM100/SM103 and Triton kernels on other NVIDIA GPUs. [Kimi K3] allow KDA using inline Attention Gym reference eager implementation to support non-SM100/103 devices #4374 was closed for this reason.K3's real hardware floor is the fused bounded gate (
bound_gate(impl="fused")), which needs TMA, i.e. CUDA capability 9.0+.Changes
KDAKernel: require CUDA capability >= 9.0 instead of exactly SM100/SM103.test_kda_attention.py,test_kimi_k3.py: relax the KDA skip conditions to match.SM100/SM103 behavior is unchanged