Skip to content

Pull requests: ROCm/composable_kernel

New pull request New
58 Open 3,332 Closed
58 Open 3,332 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[CK Tools] Auto-enable unbuffered output for Python commands
#3694 opened Jan 30, 2026 by AviralGoelAMD Loading…
[CK Conv] Add bwd weight instance for large-k shape
#3692 opened Jan 30, 2026 by johannes-graner Loading…
7 tasks
[CK_TILE] Extend support of mix precision microscaling BQuant organization: streamhpc
#3689 opened Jan 30, 2026 by EnricoDeg • Draft
7 tasks
@EnricoDeg
[CK_TILE]: PreshuffleB + PreshuffleBQuant for ABQuant pipeline organization: streamhpc
#3687 opened Jan 30, 2026 by ErwinTerpstra • Draft
6 of 7 tasks
Adapt parser to monorepo
#3686 opened Jan 30, 2026 by randyspauldingamd Loading…
1 of 7 tasks
Grouped Convolution Backward Data Direct Load
#3685 opened Jan 30, 2026 by jakpiase Loading…
Enable Grouped Conv Tile Fwd Tests daily
#3680 opened Jan 29, 2026 by bartekxk Loading…
7 tasks done
@bartekxk
[Conv] Add NumGroupsToMerge to BwdWeight type string
#3679 opened Jan 29, 2026 by johannes-graner Loading…
7 tasks
feat: add new optimized tutorial kernels
#3676 opened Jan 29, 2026 by AviralGoelAMD • Draft
7 tasks
[CK] Add fwd conv group merging to v3 conv instances
#3675 opened Jan 29, 2026 by vpietila-amd Loading…
Add padding to cshuffle epilogue to avoid bank conflict
#3671 opened Jan 28, 2026 by ThomasNing Loading…
7 tasks
11
[CK] Add new fwd conv fp16/bf16 instances optimized for unit group size.
#3670 opened Jan 28, 2026 by vpietila-amd Loading…
5
Implement fp16 x fp8, bf16 x fp8, fp8 x fp16, and fp8 x bf16 for the V3 pipeline
#3669 opened Jan 28, 2026 by SamiAario-AMD • Draft
3 of 7 tasks
173 implement device grouped gemm fixed nk for rdna4 organization: streamhpc
#3668 opened Jan 28, 2026 by bidlekm Loading…
5 of 7 tasks
Add a README.md file to ck/library/util
#3665 opened Jan 28, 2026 by shumway Loading…
15
Adding Additional Failure Patterns for Alerts
#3663 opened Jan 27, 2026 by amd-anclark Loading…
4
[CK_TILE] Stream-K Tile Engine Test Config File Generation
#3662 opened Jan 27, 2026 by ecamartins Loading…
5 of 7 tasks
@ecamartins
9
[CK] add inter/intrawave scheduling concept doc ci:docs-only Skip most non-doc CI for this PR documentation Improvements or additions to documentation
#3660 opened Jan 27, 2026 by spolifroni-amd Loading…
10
Re enable two fp16 x pkint4 tests
#3655 opened Jan 27, 2026 by SamiAario-AMD • Draft
4 of 7 tasks
test: Add umbrella test targets for CK Tile operations
#3654 opened Jan 26, 2026 by AviralGoelAMD Loading…
6 of 7 tasks
7
feat: add split_k support for block scale gemm bquant mode.
#3653 opened Jan 26, 2026 by AviralGoelAMD Loading…
7 tasks
@AviralGoelAMD
14
[CK_TILE] Stream-K XCD remapping
#3652 opened Jan 26, 2026 by arai713 • Draft
4 of 7 tasks
@arai713
5
[CK_TILE] Add blockscale GEMM support for EightWarps on gfx950
#3650 opened Jan 26, 2026 by kensclin Loading…
2 of 7 tasks
Refactor vector type to reduce build times
#3641 opened Jan 23, 2026 by cgmillette Loading…
7 tasks
9
[Compiler] Addressing new compiler warnings
#3640 opened Jan 23, 2026 by jplehr Loading…
1 of 7 tasks
3
ProTip! Add no:assignee to see everything that’s not assigned.