Skip to content

Pull requests: tile-ai/tilelang

New pull request New
36 Open 1,579 Closed
36 Open 1,579 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

[feature][Blackwell] Add SM120 FP4 and A8W4 GEMM support
#2171 opened May 8, 2026 by TerminusAkivili Contributor Loading…
6
[Refactor] Remove body from TIR LetStmt
#2170 opened May 8, 2026 by LeiWang1999 Member Loading…
9
[Example] Add CLC-pipelined 2-CTA GEMM example for sm100
#2169 opened May 8, 2026 by ighoshsubho Loading…
6
[Example] Add layernorm example in tilelang
#2168 opened May 8, 2026 by ighoshsubho Loading…
5
[BugFix] Consider non-local store in external call and SIMT producer for warp specialize
#2166 opened May 7, 2026 by Rachmanino Collaborator Loading…
4
[Refactor] Refactor multiple TensorCoreIntrinEmitter to provide atom-level mma control interface
#2161 opened May 7, 2026 by Rachmanino Collaborator Loading…
5 tasks done
12
[Autotune] Add pipeline, grouped compilation, and multi-GPU benchmark support
#2159 opened May 7, 2026 by Wazrrr Loading…
18
Fix atomic_load access_ptr lowering for dynamic indices
#2157 opened May 6, 2026 by VitalyAnkh Contributor Loading…
1
2
[feat] Add Hexagon HMX backend support
#2155 opened May 6, 2026 by Calaweh Loading…
1
18
[Metal] FP8 vector cast lanes 2/3/4 (extends storage-only FP8)
#2145 opened May 4, 2026 by apstenku123 Loading…
6 tasks done
18
[Metal] FP8 storage-only emulation (uchar storage + LUT decode helpers)
#2144 opened May 4, 2026 by apstenku123 Loading…
5 of 6 tasks
15
[Metal] emit Metal builtins directly instead of CUDA-style threadIdx/blockIdx aliases
#2143 opened May 4, 2026 by apstenku123 Loading…
3 of 4 tasks
20
tilelang: T.fp8_scaled_matmul DSL intrinsic + Metal lowering
#2142 opened May 4, 2026 by apstenku123 Loading…
16
[Metal] thread stage dim through T.access_ptr for T.Pipelined num_stages>1
#2141 opened May 4, 2026 by apstenku123 Loading…
14
[Metal] route FP8-input T.gemm to scalar fallback
#2140 opened May 4, 2026 by apstenku123 Loading…
16
[Metal] allow mixed-dtype T.gemm via scalar fallback
#2139 opened May 4, 2026 by apstenku123 Loading…
11
[Fix] Allow autotuning kernels with scalar value parameters
#2136 opened May 3, 2026 by yurekami Contributor Loading…
2 of 3 tasks
1
6
[AMD][CDNA4] Add MXFP4 (FP4 E2M1) support for gfx950
#2132 opened Apr 30, 2026 by zhangnju Collaborator Loading…
7
Rebase Metal simdgroup GEMM support and runtime coverage
#2130 opened Apr 30, 2026 by jorgecurious Loading…
24
Add TMA tile::gather4 / tile::scatter4 support
#2129 opened Apr 30, 2026 by ighoshsubho Loading…
3 of 4 tasks
26
Add RDNA gfx1151 ROCm target support
#2127 opened Apr 29, 2026 by lhl Loading…
41
Add Metal scalar fallback for T.gemm
#2118 opened Apr 29, 2026 by cklxx Loading…
9
[Feature] Support mutable TMA descriptor and canonicalize usage in examples
#2113 opened Apr 28, 2026 by Rachmanino Collaborator • Draft
2
feat: auto-vectorize bf16/fp16 reduce with packed add2 intrinsics
#2112 opened Apr 28, 2026 by kurisu6912 Collaborator Loading…
1
21
[Example] Optimize topk selector for B=1 and large S cases
#2108 opened Apr 27, 2026 by Rachmanino Collaborator • Draft
5
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.