-
Notifications
You must be signed in to change notification settings - Fork 28
Pull requests: AMD-AGI/Primus-Turbo
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
flydsl: optimize mxfp4 quant kernel part2
#474
opened Aug 21, 2026 by
kyle-256
Collaborator
Loading…
12 tasks
fix: remove m_total from flydsl compile cache key
#473
opened Aug 21, 2026 by
RuibinCheung
Collaborator
Loading…
4 of 12 tasks
Add torch-compatible flex_attention entry backed by Primus-Turbo
#472
opened Aug 21, 2026 by
botaohu001
Contributor
•
Draft
12 tasks
FlyDSL: pad-aware grouped-fp8 down-proj and native fused grad-accum for gpt-oss-20b MoE
#470
opened Aug 21, 2026 by
kyle-256
Collaborator
Loading…
6 of 12 tasks
feat(attention): add Gluon forward attention backend
ci:gpu
#469
opened Aug 21, 2026 by
bangtianliu
Loading…
7 of 12 tasks
fix: loosen flydsl install requires on gfx1250
ci:gpu
#467
opened Aug 20, 2026 by
RuibinCheung
Collaborator
Loading…
5 of 12 tasks
feat(moe): support HIP and Triton backends for moe_permute, default to Triton
#455
opened Aug 12, 2026 by
zhenhuang12
Collaborator
Loading…
10 of 12 tasks
[WIP]feat(tuning): offline tune framework and configs
#439
opened Jul 30, 2026 by
xiaobochen-amd
Collaborator
Loading…
12 tasks
feat: refine quant config arguments
ci:gpu
#379
opened Jun 12, 2026 by
RuibinCheung
Collaborator
Loading…
3 of 12 tasks
feat: add more activation func
#329
opened May 7, 2026 by
RuibinCheung
Collaborator
Loading…
8 of 9 tasks
opt(gemm): add hipBLASLt algorithm cache and thread-local workspace
#321
opened Apr 30, 2026 by
jasainio
Contributor
Loading…
6 of 12 tasks
Refactor: moe dispatch combine autotune
ci:gpu
#312
opened Apr 24, 2026 by
zhenhuang12
Collaborator
Loading…
7 of 12 tasks
feat: enable hybrid FP8 dtypes on Triton grouped GEMM backends
#288
opened Apr 15, 2026 by
sarthak-amd
•
Draft
refactor: reorganize moe ops and kernels
#243
opened Mar 5, 2026 by
zhenhuang12
Collaborator
Loading…
ProTip!
Mix and match filters to narrow down what you’re looking for.