-
Notifications
You must be signed in to change notification settings - Fork 732
Pull requests: tile-ai/tilelang
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
[CUDA] Support missing 16-bit inverse and binary math operations
#3172
opened Sep 6, 2026 by
Chennesxu
Contributor
Loading…
[Layout] Consider scalar reducer plans in register-count search
#3171
opened Sep 6, 2026 by
LeiWang1999
Member
Loading…
[ROCm] Run portable example validation in CI
#3165
opened Sep 3, 2026 by
andyluo7
Collaborator
Loading…
[ROCm] Validate sparse MLA backward with configurable launch tiles
#3159
opened Sep 3, 2026 by
andyluo7
Collaborator
Loading…
[ROCm] Support DeepSeek mHC pre and post kernels
#3156
opened Sep 3, 2026 by
andyluo7
Collaborator
Loading…
[ROCm] Support DeepSeek NSA forward and decode
#3152
opened Sep 3, 2026 by
andyluo7
Collaborator
Loading…
[ROCm] Validate linear attention forward and backward
#3151
opened Sep 3, 2026 by
andyluo7
Collaborator
Loading…
[ROCm] Select feasible flash-decoding tiles
#3149
opened Sep 3, 2026 by
andyluo7
Collaborator
Loading…
[Bugfix][Quantize] Use plain int masks in interleave_weight special branches
#3140
opened Sep 2, 2026 by
rishabhsinha17
Contributor
Loading…
[BugFix] Split divergent branches around required syncs
#3138
opened Sep 2, 2026 by
haoyang9804
Contributor
Loading…
[BugFix] Allow async copies in conditionally executed pipeline stages
#3137
opened Sep 2, 2026 by
zhouyangye1076
Contributor
Loading…
[Lang] Add T.auto_alloc with automatic memory-scope inference
#3130
opened Sep 1, 2026 by
penguin-wwy
Contributor
Loading…
[BugFix] Coalesce sub-32-bit elements in parallel loop partition
#3127
opened Sep 1, 2026 by
zhouyangye1076
Contributor
Loading…
[BugFix] Reject non-bijective sm8x sparse metadata shapes
#3125
opened Sep 1, 2026 by
saiyambharara
Loading…
1 task done
[CPU] Support OpenMP parallel of grid loops
#3109
opened Aug 28, 2026 by
penguin-wwy
Contributor
Loading…
[ROCm] Support packed INT4x2/UINT4x2 codegen
#3103
opened Aug 27, 2026 by
andyluo7
Collaborator
Loading…
[SM120] kind::mxf8f6f4 block-scaled MMA: full f8f6f4 operand family, sub-byte TMA producers, and official examples
#3099
opened Aug 27, 2026 by
qqq-tao
Contributor
Loading…
[CUDA] Add SM120 mxf4nvf4 2X/ue8m0 and 4X/ue8m0 block-scale MMA support
#3081
opened Aug 25, 2026 by
qqq-tao
Contributor
Loading…
[BugFix][Transform] Clamp short symbolic software-pipeline epilogues
#3072
opened Aug 23, 2026 by
3402956340
Loading…
[BugFix][Layout] Accept Python int coalesced width annotations
#3064
opened Aug 22, 2026 by
kobecai
Contributor
Loading…
6 tasks done
[BugFix][Layout] Fix parallel layout undercoverage
#3063
opened Aug 21, 2026 by
haoyang9804
Contributor
Loading…
Previous Next
ProTip!
What’s not been updated in a month: updated:<2026-08-06.