PyPI · #2341 most downloaded on PyPI

tilelang

0.1.153.0M downloads/motile-ai/tilelang ↗New releases in your daily digest, changelog included.

A tile level programming language to generate high performance code.

Last release 5 days ago
30 Sep 2026
Ships fairly regularly
a new release about every 4 weeks
Nearly every release is documented
notes for 22 of 22 stable releases
1 version withdrawn
withdrawn after publishing
2 years old
24 releases · first in 2025

Release timeline

24 releases since 2025

20252026

One column per month.

Releases

0.1.6.post231 Oct 2025Deprecated
[Enhancement] Deprecate split&sum in attn bwd examples on Hopper by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1065
The Last Release for Python 3.8 (without tvm-ffi) 🚀

What's Changed

[Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/856

[Doc] Optimize the quickstart guide for clarity and not just for CUDA by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/858

[TMA] Bugfix when a shared buffer is both issued with tma store and tma load by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/857

[AMD][MLA] Fix mla autotune for rocm by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/861

[Bugfix] Ensure correct handling for cases where seq_q<seq_kv in flash attention examples by @Rachmanino in https://github.com/tile-ai/tilelang/pull/864

[AMD] refactor MatrixCoreIntrinEmitter by @Paran0idy in https://github.com/tile-ai/tilelang/pull/860

[Feat] Add fast sine and cosine definitions in CUDA templates by @Rachmanino in https://github.com/tile-ai/tilelang/pull/865

[Layout] Support layout forward with multi dimension by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/867

[Autotune][Conv] optimize convolution examples to use autotune by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/866

[Example] Add examples to support efficient attention sink forward process by @Rachmanino in https://github.com/tile-ai/tilelang/pull/853

[Parser] Adapt Parser to work with Python 3.8 in some cases by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/869

[Fix] Fix bug 0905: tilelang doesn't vectorize B[i,j] = c[i] + A[i,j] by @kurisu6912 in https://github.com/tile-ai/tilelang/pull/798

[Language] Support sequence comparisons by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/872

[Language] Support loop_break primitive by @chengyupku in https://github.com/tile-ai/tilelang/pull/873

[Bugfix] Use ExprDeepEqual instead of StructuralEqual when merge consecutive If stmt by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/876

[Language] Support atomic add with ret by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/870

[Cython] Remove an incorrect check by @LJC00118 in https://github.com/tile-ai/tilelang/pull/880

Update amd_ci.yml by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/881

[FastMath] Disable default TVM fastmath intrinsic dispatch and add explicit fastmath op to invoke by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/875

[Example] Add efficient attention sink backward implementations and tests by @Rachmanino in https://github.com/tile-ai/tilelang/pull/877

[Precision] Introduce T.ieee_rsqrt and related high precision op by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/882

[Dist] Provide an option to include commit ID in version by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/884

[Example] Optimize sink attention forward via swizzled layout and report benchmark results by @Rachmanino in https://github.com/tile-ai/tilelang/pull/885

[Layout] Introduce Flexible Parallel to Support T.serial and local buffers inside T.Parallel loop by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/844

[Bugfix][Enhancement] Fix a bug in previous commit and enhance cuda backend by @Hamerlate in https://github.com/tile-ai/tilelang/pull/887

[Bugfix] Fix CopyNode Lower method to include disable_tma flag in GetCopyInst by @Rachmanino in https://github.com/tile-ai/tilelang/pull/888

[Layout] Fix plot layout by @Paran0idy in https://github.com/tile-ai/tilelang/pull/890

[Example] Add example by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/894

[News] Add announcement of support for Huawei Ascend chips by @xwhzz in https://github.com/tile-ai/tilelang/pull/895

[Example] Add sparse mla examples by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/896

[Typo] Fix backend name for Huawei Ascend by @xwhzz in https://github.com/tile-ai/tilelang/pull/898

[CI] Legalize math related test by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/899

[Bugfix] Fix flops comp and softmax scale in mla by @Edenzzzz in https://github.com/tile-ai/tilelang/pull/900

[Example] Specify a fixed commit for the flash-linear-attention repository and optimize nsa examples by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/913

[CI] optimize CI time for sparse gemm by @botbw in https://github.com/tile-ai/tilelang/pull/906

[Enhancement] Include compile flags into the hash key of cached kernels by @Rachmanino in https://github.com/tile-ai/tilelang/pull/911

[Bugfix] Fix saving kernel source code where JITKernel.artifact is None by @zjudmd1015 in https://github.com/tile-ai/tilelang/pull/921

[CI] Refactor import paths in dequantization examples to use dequantize_utils by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/914

[Example] Add MLA decode ws example by @chengyupku in https://github.com/tile-ai/tilelang/pull/928

[CI] Fix documentation runner by adding 'nvidia' tag by @xwhzz in https://github.com/tile-ai/tilelang/pull/927

[Layout] Strict annotate completed replicated layout for fragment with constant index by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/929

[Bugfix] Fix tensor memory copy layout by @Hamerlate in https://github.com/tile-ai/tilelang/pull/933

[Example] Optimize online_softmax example by @lijinpei in https://github.com/tile-ai/tilelang/pull/934

[Example] Add correctness assert into dsa example by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/937

[Enhancement] Enhance and add new GQA backward examples for Hopper by @Rachmanino in https://github.com/tile-ai/tilelang/pull/930

[Enhancement] Fix lint to improve grouped GEMM performance with TMA by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/938

[Example] Introduce split+sum template, and optimize atomic_add performance for bwd examples by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/940

[Example] Disable TMA and enable FastMath for NSA Examples (#941) by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/941

[Example] Revert the atomic/split&sum templates in MHA backward examples by @Rachmanino in https://github.com/tile-ai/tilelang/pull/943

[Example] Add sparse mla bwd example for deepseek_v32 by @Zhichenzzz in https://github.com/tile-ai/tilelang/pull/919

[Profiler]Adds CUPTI profiler support by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/936

[Enhancement] Support Copy for Buffer Load witih scalar indices by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/946

[Code Style] Refine nvrtc compile related check style by @BBuf in https://github.com/tile-ai/tilelang/pull/945

[Backend] Add metal backend by @oraluben in https://github.com/tile-ai/tilelang/pull/799

[CI] enable dependabot for GHA workflows by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/950

Modify the SM architecture number to support Thor’s sm110. by @iloveai8086 in https://github.com/tile-ai/tilelang/pull/957

[CI] auto-cancel in-progress PR CI when new commits are pushed by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/956

[bug] fix type object is not subscriptable in py38 by @BBuf in https://github.com/tile-ai/tilelang/pull/959

[Bugfix][Doc] Add astroid version constraint to requirements.txt by @xwhzz in https://github.com/tile-ai/tilelang/pull/958

[CI]: Bump actions/setup-python from 2 to 6 by @dependabot[bot] in https://github.com/tile-ai/tilelang/pull/951

[CI]: Bump astral-sh/setup-uv from 6 to 7 by @dependabot[bot] in https://github.com/tile-ai/tilelang/pull/952

[CI]: Bump actions/github-script from 7 to 8 by @dependabot[bot] in https://github.com/tile-ai/tilelang/pull/954

[CI]: Bump actions/checkout from 2 to 5 by @dependabot[bot] in https://github.com/tile-ai/tilelang/pull/953

[TileOp] Implement WGMMA for T.gemm_v2 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/813

[Docs] add CODE_OF_CONDUCT.md by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/965

[Example] Add support for bfloat16 and user-defined sm_scale in attention sink examples by @Rachmanino in https://github.com/tile-ai/tilelang/pull/924

[Bugfix] Do not force inline let stmt by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/947

[CI] add pre-commit integration by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/955

[Doc] Install docs add docker install method by @BBuf in https://github.com/tile-ai/tilelang/pull/961

[Bugfix] Fix dummy kernel compliation by @SiriusNEO in https://github.com/tile-ai/tilelang/pull/962

[CI][Refactor] Refactor non-test CI workflow files by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/971

[TileOp] Implememt CumSum1D by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/978

[Language] Enhance T.alloc_var for AugAssign and AnnAsign by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/979

[Refactor] Refactor Pass InjectFenceProxy and expose some warp group primitives in frontend by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/977

[Typo] Remove debug print by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/980

[Bugfix] Use access_ptr("r") instead of access_ptr("w") for correct pipeline analysis by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/983

[Feature][Example] Support TMA reduce operation and update GQA bwd example by @chengyupku in https://github.com/tile-ai/tilelang/pull/969

[Bugfix] Add NVIDIA HPC SDK support in CUDA detection (#974) by @Degeneracy-Evil in https://github.com/tile-ai/tilelang/pull/976

[BugFix] Robust gemm policy for sparse_mla_fwd in Hopper and Ada Lovelace architectures by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/984

[Bugfix] Fallback torch.accelerator.synchronize() to torch.cuda.synchronize() by @yyttt6 in https://github.com/tile-ai/tilelang/pull/987

[Bugfix]:Fix atomicadd auto vectorize identify var error by @yyttt6 in https://github.com/tile-ai/tilelang/pull/883

[CI] Speed up sparse tensor core test via vectorized generating sparse data by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1009

[Build] Migrate to scikit-build-core by @oraluben in https://github.com/tile-ai/tilelang/pull/939

[CI] Removes redundant environment variable by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/1020

[Transform] Migrate LowerIntrin from tvm into tilelang by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/999

[Lint] Prefer American English spelling by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/1022

[Build] Prefer libs from local build dir by @oraluben in https://github.com/tile-ai/tilelang/pull/1027

[Language] Support Consequential assignments like 'a = b = c = 1' by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/992

[CI] Removes debug print statements from the example. by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/1030

[Enhancement] Update abs function for half_t and bfloat_t to use cutlass implementation by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1023

[Bugfix] Recover code for flexible parallel by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1032

[CI] Disable buggy(maybe) warp specialized kernel ci test for H20 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1033

[TIR] Revert some changes of Pass LowerIntrin by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1035

[Env] Optimize the mechanism for locating TL_LIBS by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1038

[CUDA] Add pack functions for FP8 types by @LJC00118 in https://github.com/tile-ai/tilelang/pull/967

[Language] Expose T.get_warp_idx_sync and T.shuffle_elect for efficient thread election by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/989

[AMD] fix bug&add amd fp8 examples by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/966

[CI][Refactor] Merge test CI workflow files into one by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/973

[BugFix] Phaseout dependency of Triton in sink examples to make CI happy by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1045

[Refactor] Use has_simt_copy to decide whether to insert set_max_nreg by @chengyupku in https://github.com/tile-ai/tilelang/pull/982

[Feature]: Add test for atomicadd auto vectorize and remove useless code by @yyttt6 in https://github.com/tile-ai/tilelang/pull/1019

Allow mma gemm for all cuda arch by @oraluben in https://github.com/tile-ai/tilelang/pull/1047

[Bugfix] Improves compatibility when checking for MPS availability in different PyTorch builds. by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1051

[CI] Fix ROCm CI by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/1043

[Enhancement] Add support for symbolic dimensions in Cython kernel adapter and improve static shape validation in wrapper by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1024

Automatically initialize submodule if missing by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1052

[Enhancement] Remove constraint requiring last dimension stride to be 1 by @LJC00118 in https://github.com/tile-ai/tilelang/pull/1040

[CI] Disable autofix for pre-commit CI by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1053

[Enhancement] Improve CUDA compiler detection in CMake by @LJC00118 in https://github.com/tile-ai/tilelang/pull/1054

[Enhancement] Introduce a workaround for layout inference for local buffer store by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1055

[Refactor] Refactor Pass LegalizeSafeMemoryAccess to support recursive load/store rewrite by @SiriusNEO in https://github.com/tile-ai/tilelang/pull/1050

Making version parser more robust against missing or unavailable metadata by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1061

[DOC] Add document for develop with PYTHONPATH by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1062

[CI]:Reduce test shapes to avoid OOM errors during CI. by @yyttt6 in https://github.com/tile-ai/tilelang/pull/1060

[Benchmark] Add H800 SXM Benchmark results by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1063

[Misc] Add GitHub issue templates by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/1057

[Refactor][Example] Update linear attention examples and add tests by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1010

[Enhancement] Deprecate split&sum in attn bwd examples on Hopper by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1065

[Benchmark] Add matmul FP16 benchmark results by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1067

[CI]: Bump actions/checkout from 4 to 5 by @dependabot[bot] in https://github.com/tile-ai/tilelang/pull/1070

[Example] Update GQA varlen fwd and MHA varlen fwd by @chengyupku in https://github.com/tile-ai/tilelang/pull/1071

[Parallel] Support T.Parallel with dynamic extents by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/990

[Layout] Utilizing IsEqual instead of StructuralEqual by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1073

[Cache] raise errors for tileang.clear_cache() by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1077

[Feature] Support Reduce operators for bitwise and/or/xor by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/1074

[Autotune] Add autotune coverage for symbolic M and normalize cache key by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1075

[Language] Recommend using T.dynamic instead of T.symbolic by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1076

[Language] Efficient T.reduce_ with shared memory input/output by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1080

[Bugfix] Fix missing reg alloc in custom warp specialization by @chengyupku in https://github.com/tile-ai/tilelang/pull/1084

[Enhancement] Update async intrinsic handling in inject_fence_proxy by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1068

[Feature] Add GQA backward kernel with varlen input by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/1082

[BugFix] Add memory order argument for non-vectorized atomic add by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/1081

[Refactor] Rename cython output to tilelang_cython and relocate its path by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1086

[Target] Enhance target selection helpers and documentation by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1085

[Cleanup] Remove tilelang.disable_cache() calls from examples and tests by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1088

[PassConfig] Introduce PassConfig TL_STORAGE_REWRITE_DETECT_INPLACE by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1089

[Language] Support tilelang alloc_var(dtype, init=x) by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1092

[Bugfix] Fix missing host cuTensorMapEncodeIm2col call by @chengyupku in https://github.com/tile-ai/tilelang/pull/1094

[GQA] Add regional atomic add to slightly boost performance by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/1093

[Example] Add block level high performance gemv example by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1097

[Refactor] Optimize debug message for parallel inference by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1096

[CI][Lint] Retire format.sh and add clang-tidy to GHA workflow by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/1044

[Refactor] Use forceinline in ldmatrix and update mamba scan kernel by @chengyupku in https://github.com/tile-ai/tilelang/pull/1104

[Maint] Update uncommitted change detection command in format.sh by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/1102

[Benchmark] Add Mamba2_chunk_scan benchmark by @chengyupku in https://github.com/tile-ai/tilelang/pull/1109

[Benchmark] Update Mamba2_chunk_scan benchmark by @chengyupku in https://github.com/tile-ai/tilelang/pull/1110

[Lint] Enable pyupgrade linter in ruff by @oraluben in https://github.com/tile-ai/tilelang/pull/963

[Refactor] Improve scalar handling in CopyNode and update loop partition dtype logi by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1111

[Feature] Enhance vectorized conversion support in CUDA codegen by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1095

[Feature] Support None type as input for T.ptr and T.Tensor by @xwhzz in https://github.com/tile-ai/tilelang/pull/1114

[Bugfix] Resolve mixed stride dtype issue (inconsistent int32/int64 values) by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1119

[Feature] Add memory_order PTX for vectorized atomic add by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/1112

[CI]: Bump actions/upload-artifact from 4 to 5 by @dependabot[bot] in https://github.com/tile-ai/tilelang/pull/1128

[CI]: Bump actions/download-artifact from 5 to 6 by @dependabot[bot] in https://github.com/tile-ai/tilelang/pull/1127

[Enhancement] Add missing fence_barrier_init primitive after mbarrier init by @chengyupku in https://github.com/tile-ai/tilelang/pull/1121

[Feature]:Add device assert by @yyttt6 in https://github.com/tile-ai/tilelang/pull/1116

[Build][CI] Build and test SDist in release CI by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/1098

[Benchmark] Update triton and helion baselines in mamba-chuk-scan by @chengyupku in https://github.com/tile-ai/tilelang/pull/1131

Add int2 and longlong4 pack functions by @LJC00118 in https://github.com/tile-ai/tilelang/pull/1129

[BugFix] Add memory order and testing script for split version GQA bwd kernel by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/1100

[Bugfix] Correctly construct the argument list for atomic add based on the vector size by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1137

[AMD] Supoort T.gemm_v2 for AMD Backend by @Paran0idy in https://github.com/tile-ai/tilelang/pull/1136

[BugFix] alloc_var init failed to handle complex expression by @kurisu6912 in https://github.com/tile-ai/tilelang/pull/1144

[Refactor] Remove amd gemm_v2 tests by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1149

[BugFix] Implement bfloat16 support in CUDA code generation with min/max functions and inf/nan values by @Rachmanino in https://github.com/tile-ai/tilelang/pull/1143

[Bugfix] Implement classic arena algorithm for shmem merge and WAW conflict detection by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1146

[CI] allow dirty workspace for format.sh and introduce loop carry thread sync unit test by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1153

[CI] use Python urllib to download file instead of Wget by @XuehaiPan in https://github.com/tile-ai/tilelang/pull/1154

[BugFix] Correct direct copy from bf16 to fp8 by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/1090

[Refactor]:Move device_assert from extern_call to intrin_call by @yyttt6 in https://github.com/tile-ai/tilelang/pull/1134

[Enhancement] Enhance Cast operations Vectorization by @LJC00118 in https://github.com/tile-ai/tilelang/pull/1156

[Bugfix] Enhance LetStmt handling in Vectorize Loop Pass by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1159

[Release] Bump version to v0.1.6.post2 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/1160

New Contributors

@LJC00118 made their first contribution in https://github.com/tile-ai/tilelang/pull/880

@Edenzzzz made their first contribution in https://github.com/tile-ai/tilelang/pull/900

@zjudmd1015 made their first contribution in https://github.com/tile-ai/tilelang/pull/921

@lijinpei made their first contribution in https://github.com/tile-ai/tilelang/pull/934

@Zhichenzzz made their first contribution in https://github.com/tile-ai/tilelang/pull/919

@BBuf made their first contribution in https://github.com/tile-ai/tilelang/pull/945

@XuehaiPan made their first contribution in https://github.com/tile-ai/tilelang/pull/950

@iloveai8086 made their first contribution in https://github.com/tile-ai/tilelang/pull/957

@Degeneracy-Evil made their first contribution in https://github.com/tile-ai/tilelang/pull/976

Full Changelog: https://github.com/tile-ai/tilelang/compare/v0.1.6.post1...v0.1.6.post2

0.1.619 Sept 2025withdrawn: static link g++

[Bugfix] Added missing thread offsets and other information to reduce by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/646

What's Changed

[Bugfix] Added missing thread offsets and other information to reduce by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/646
[Bugfix] Adjust role assignment in warp specialization based on read access by @chengyupku in https://github.com/tile-ai/tilelang/pull/647
Fix/jit kernel use target by @meinie0826 in https://github.com/tile-ai/tilelang/pull/648
[Bugfix] Remove small array reuse condition in shared memory allocation merging by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/654
[Enhancement] Add role assignment for AllocateNode in warp specialization by @chengyupku in https://github.com/tile-ai/tilelang/pull/657
[Bugfix][CI] Bug fixing and migrate CI from ada to hopper by @xwhzz in https://github.com/tile-ai/tilelang/pull/652
[CI] Enable cache for virtual env and parallelize pytest via xdist by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/660
[Cache] Support shared cache directories for multiple process by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/649
[Enhancement] Add compile_flags parameter to JIT kernel and adapter classes for improved compilation control by @xwhzz in https://github.com/tile-ai/tilelang/pull/656
add the support of rocm arch detecting by @zhangnju in https://github.com/tile-ai/tilelang/pull/661
[BugFix] Do not modify strict layout in common or relax level of layout inference. More conditions on layout checking by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/653
[Bugfix][Docs] Update documentation build process and configurations for autoapi support by @xwhzz in https://github.com/tile-ai/tilelang/pull/663
[Enhancement] Improve buffer conflict detection in thread storage synchronization by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/658
[Bugfix] Consider buffer data type into indices provably disjoint analysis by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/664
[Bugfix] Remove redundant T.fill to fix precision issue by @xuchangtolearn in https://github.com/tile-ai/tilelang/pull/667
[Enhancement] Refactor buffer index handling for improved precision a… by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/671
Reverts tile-ai/tilelang#671 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/672
[Bugfix] Passing correct nvcc to cmake by @chenyang78 in https://github.com/tile-ai/tilelang/pull/670
[CI] Improve format check output and automate commit of changes by @xwhzz in https://github.com/tile-ai/tilelang/pull/669
[Bugfix][CI] Use valid runner labels in workflow by @xwhzz in https://github.com/tile-ai/tilelang/pull/674
[Enhancement] passing verbose to LibraryGenerator by @chenyang78 in https://github.com/tile-ai/tilelang/pull/673
[Enhancement] Enhance lint error messaging in CI by @xwhzz in https://github.com/tile-ai/tilelang/pull/675
Refactor to support upstream tvm by @Hzfengsy in https://github.com/tile-ai/tilelang/pull/595
Do not check for short variables by @oraluben in https://github.com/tile-ai/tilelang/pull/676
[Refactor] Phaseout version with commit id in editable model by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/677
[CI] Update CI workflow to use Python 3.12 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/679
[Enhancement] Output cache-file-related messages with verbose=True by @chenyang78 in https://github.com/tile-ai/tilelang/pull/683
[Enhancement] Enhance warp specialization logic by @chengyupku in https://github.com/tile-ai/tilelang/pull/680
Add Flash Attn example on amd mi300 series by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/682
[Enhancement] Refactored buffer detection logic in warp_specialized_rewriter.cc by @chengyupku in https://github.com/tile-ai/tilelang/pull/685
[Fix] fix some issues with JIT decorators existing in the examples by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/681
[Enhancement] Add --ptxas-options=--register-usage-level=10 option by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/684
[Feature]:Add auto vectorize for atomic add by @yyttt6 in https://github.com/tile-ai/tilelang/pull/686
[Refactor] Rebase pipeline injector from upstream tvm by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/687
[Refactor] Introduce GemmInst for different targets handling by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/688
[Enhancement] Optimize BF16 casting performance by @xwhzz in https://github.com/tile-ai/tilelang/pull/689
[Smem Reuse] Optimize to do memory alignment on identical buffers. by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/693
[Version] Keep local commit id as it somehow help with debugging by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/697
[Example] Optimize warp specialize flashmla example by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/698
Bump transformers from 4.52.1 to 4.53.0 in /examples/bitnet-1.58b by @dependabot[bot] in https://github.com/tile-ai/tilelang/pull/700
Gated Delta Net(GDN) kernel implementation in TileLang by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/695
Trivial update to calculate target arch by @oraluben in https://github.com/tile-ai/tilelang/pull/702
[CI] Remove Flash Attention dependency by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/705
[Layout] Introduce a new layout inference mechanism by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/699
[Pipeline] Optimize inject software pipeline and pipeline planing pass by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/706
Low-bit kernels fix and implementation by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/704
[Feat] Support gemm with stride by @smallscientist1 in https://github.com/tile-ai/tilelang/pull/701
[Enhancement] Add eviction policy support for TMA operations, enhance CUDA codegen, and introduce new pass config by @xwhzz in https://github.com/tile-ai/tilelang/pull/690
[Enhancement] Enhance the robustness and generality of MLA examples by @Rachmanino in https://github.com/tile-ai/tilelang/pull/709
[Refactor] MergeAnnotations function to accept Map<Any, Any> instead of Map<String, Any> by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/710
[Pipeline] Phaseout fragment and double buffer info from pipeline pass by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/711
[Pipeline] Skip condition expression analysis for global reading by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/713
[Index] Relocate Int64 Auto Promoter to ConfigBitWidth Pass, removing it from FlattenBuffer by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/714
[CI] Bind build-test CI to NVIDIA as AMD runners are being introduced by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/718
fix: NVRTC backend by @lucifer1004 in https://github.com/tile-ai/tilelang/pull/717
[CUDA] Init support for sm_120 by @oraluben in https://github.com/tile-ai/tilelang/pull/716
[Bugfix] Correct git configuration in docs CI by @xwhzz in https://github.com/tile-ai/tilelang/pull/720
[Chore] fix typos by @lucifer1004 in https://github.com/tile-ai/tilelang/pull/719
[CI][AMD] Add AMD GPU CI and fix some related bugs by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/694
[Carver][Bugfix] Correct score function for warp tile selection in tensorcore policy by @NaOHCC in https://github.com/tile-ai/tilelang/pull/724
[Refactor] Refactor CUDA code generation to simplify eviction policy handling by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/721
[Language] Introduce StridedTensor to support non contigious torch inputs by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/722
[Enhancement][Bugfix] Fix bug in warp specialized pass and add gemm_sr fallback support for Hopper by @xwhzz in https://github.com/tile-ai/tilelang/pull/712
📝 Add docstrings to fix by @coderabbitai[bot] in https://github.com/tile-ai/tilelang/pull/726
fix amd ci&add examples by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/729
[Feature] Low-bit twiddling dequantization and FP4 GEMM by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/725
📝 Add docstrings to mxfp4 by @coderabbitai[bot] in https://github.com/tile-ai/tilelang/pull/732
[Refactor] Refactor env into a more flexible version by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/740
[Bugfix] Align stride index validation with torch in CythonKernelWrapper by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/743
[Bugfix]:Fix atomic add auto vectorize memory access out of bound error by @yyttt6 in https://github.com/tile-ai/tilelang/pull/742
📝 Add docstrings to main by @coderabbitai[bot] in https://github.com/tile-ai/tilelang/pull/745
[Refactor] Refactor barrier management by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/744
[Refactor] Merge bulk copy into copy and improve layout inference for bulk copy by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/746
[Refactor] Merge ThreadPartialSync and ThreadStorageSync by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/741
[Enhancement] Optimize loop body handling in IR by @chengyupku in https://github.com/tile-ai/tilelang/pull/749
[MXFP4] Fix bugs and optimize exponential operation by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/750
[Enhancement] Add DispatchInstruction specialization for fp8 types in gemm_sm90.h by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/751
[Enhancement] Add shape checking for reduce options by @kurisu6912 in https://github.com/tile-ai/tilelang/pull/748
[Bugfix] Add missing FP8 header include by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/752
[MXFP4] Add bias to MXFP4 GEMM kernel by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/753
[Bugfix][WS] Consider loop min extent when computing phase id by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/754
[Typo] Remove disable_cache in some tests by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/755
[README] Update GDN README for clarity and add acknowledgements by @chengyupku in https://github.com/tile-ai/tilelang/pull/758
cutlass v4.2.0 supporting cuda 13 by @johnnynunez in https://github.com/tile-ai/tilelang/pull/760
[Feature] Add 1D TMA support by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/761
[Example] Add vertical slash sparse attention pattern by @xwhzz in https://github.com/tile-ai/tilelang/pull/762
[Bugfix] Address PassContext contamination from CI and fix incorrect rewrites in warp specialized pass by @xwhzz in https://github.com/tile-ai/tilelang/pull/767
[MXFP4] Add 1D TMA copy for Scale tensor in MXFP4 GEMM by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/766
[CUTLASS] hot fix blackwell by @johnnynunez in https://github.com/tile-ai/tilelang/pull/768
[Refactor] Refactor Operator into TileOperator and with tvm reflection by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/763
[Reducer] Introduce alloc_reducer to separate inter and intra warp reduction by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/757
📝 Add docstrings to pytile_0826 by @coderabbitai[bot] in https://github.com/tile-ai/tilelang/pull/770
[Bugfix]:Fix atomic add auto vectorize negative optimization by @yyttt6 in https://github.com/tile-ai/tilelang/pull/765
📝 Add docstrings to reducer_0825 by @coderabbitai[bot] in https://github.com/tile-ai/tilelang/pull/772
Allow fill global buffer by @kurisu6912 in https://github.com/tile-ai/tilelang/pull/774
[BugFix] Refactor the op check in LowerTileOp pass using the member function instead of string match by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/771
[Enhancement] Add exp fallback for bf16 by @xwhzz in https://github.com/tile-ai/tilelang/pull/776
[Lint] Introduce clang-tidy into format.sh by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/777
[Cache] Introduce detailed target information for the disk kernel cache by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/780
[Example]Adds example for top-k operation by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/775
[Math] Dispatch T.rsqrt(x) into cuda intrin instead of 1 / T.sqrt(x) by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/781
[CI] Adds pytest-durations for test timing by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/782
[Refactor] Support python reflection for tile operators by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/783
fix amd tir&add examples by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/784
[Nvidia][SM121] Add intrin.h include to gemm_mma.h for sm120+ by @HaoKang-Timmy in https://github.com/tile-ai/tilelang/pull/785
[Feat] Add tilelang T.assume support and assume injection for buffer shapes by @kurisu6912 in https://github.com/tile-ai/tilelang/pull/787
[Bugfix] Fix incorrect synchronization bug in minference example by @xwhzz in https://github.com/tile-ai/tilelang/pull/786
[AMD] fix bugs in warp shuffle by @txs19991 in https://github.com/tile-ai/tilelang/pull/790
[AMD] fix mfma op interface by @Paran0idy in https://github.com/tile-ai/tilelang/pull/791
[TMA] Automatically lower 1d tma in appropriate cases by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/788
[CI]Adds pytest timeout to CI by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/792
[Enhancement] Resolve reference cycle by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/795
[Bugfix] Fix index handling to promote 64-bit integers by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/796
[AMD] support mfma i32_16x16x32_i8 by @Paran0idy in https://github.com/tile-ai/tilelang/pull/800
[TileOp] Introduce a experimental python defined T.gemm_v2 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/793
[Bugfix] Expose alloc_reducer definition to the python side by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/802
[Refactor] Use new namespace and enhance dispatch macros for mma by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/801
[AMD] support fp8 T.gemm by @txs19991 in https://github.com/tile-ai/tilelang/pull/804
[AMD] support preshuffle weight mfma by @Paran0idy in https://github.com/tile-ai/tilelang/pull/806
Add pytest-durations to requirements for ROCm by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/810
Add ruff config to check for useless spaces by @oraluben in https://github.com/tile-ai/tilelang/pull/807
[Feature] Add ptx_cp_async_barrier_noinc intrinsic and related functionality by @chengyupku in https://github.com/tile-ai/tilelang/pull/809
[Fix] Fix lower bug when buffer store is not guarded by any tile op by @kurisu6912 in https://github.com/tile-ai/tilelang/pull/794
[feat] support gemm_sp for ampere arch by @botbw in https://github.com/tile-ai/tilelang/pull/691
[Refactor] Update TVM subproject and refactor BlockNode handling in warp_specialized_rewriter.cc by @chengyupku in https://github.com/tile-ai/tilelang/pull/812
[Refactor] Reopen #794 Fix lower bug when buffer store is not guarded by any tile op by @kurisu6912 in https://github.com/tile-ai/tilelang/pull/817
[Refactor] Update TVM subproject and streamline buffer store handling by @chengyupku in https://github.com/tile-ai/tilelang/pull/816
[Example] add w4a8 gemm kernel by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/815
[CI] fix rocm ci by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/819
[example] fix unused param in mhs example by @botbw in https://github.com/tile-ai/tilelang/pull/821
[DSL] Support python tenary if then else expression by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/822
[Bugfix] Bug fix when git command is not installed by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/823
[Bugfix] Skip fp4 dtype binding when using older versions of ml_dtypes by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/824
[Enhancement] Add a MXFP4 grouped GEMM example for FusedMoE by @Rachmanino in https://github.com/tile-ai/tilelang/pull/811
[CMake] Added support for statically linked system libc library by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/825
[Refactor] Refactor some build related configurations by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/827
[CI] Test Fix: Handle BufferLoad nodes when T.gemm input has a stride by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/843
[Refactor] Turn off ENABLE_FAST_MATH by default by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/846
[AMD] fix bf16x2 dtype codegen by @Paran0idy in https://github.com/tile-ai/tilelang/pull/847
[Typing] Fallback from Python 3.10+ type syntax for compatibility by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/848
[TIR] Refactor division simplification in RewriteSimplifier by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/849
[Py38] Revert typing and parser updates for Python 3.8 compatibility by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/850
[Bugfix] Disable Memory Info Analysis for local.var by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/851
[Release] Bump Version to 0.1.6 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/818

New Contributors

@meinie0826 made their first contribution in https://github.com/tile-ai/tilelang/pull/648
@zhangnju made their first contribution in https://github.com/tile-ai/tilelang/pull/661
@chenyang78 made their first contribution in https://github.com/tile-ai/tilelang/pull/670
@Hzfengsy made their first contribution in https://github.com/tile-ai/tilelang/pull/595
@coderabbitai[bot] made their first contribution in https://github.com/tile-ai/tilelang/pull/726
@kurisu6912 made their first contribution in https://github.com/tile-ai/tilelang/pull/748
@johnnynunez made their first contribution in https://github.com/tile-ai/tilelang/pull/760
@HaoKang-Timmy made their first contribution in https://github.com/tile-ai/tilelang/pull/785
@txs19991 made their first contribution in https://github.com/tile-ai/tilelang/pull/790
@Paran0idy made their first contribution in https://github.com/tile-ai/tilelang/pull/791

Full Changelog: https://github.com/tile-ai/tilelang/commits/0.1.6

Open source →

0.1.55 Jun 2025

[Release] Bump version from 0.1.3 into 0.1.4 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/375

What's Changed

[Release] Bump version from 0.1.3 into 0.1.4 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/375
[Enhancement] Remove redundant recursive rewrite rule for FloorDiv in RewriteSimplifier by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/408
[Docker] cu128 Support by @andyluo03 in https://github.com/tile-ai/tilelang/pull/410
[Refactor] Phaseout python dependency attrs and decorator by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/411
[Language] make linter and type checker happy with mocking by @YouJiacheng in https://github.com/tile-ai/tilelang/pull/407
[Bugfix] Support larger than 256 box size tma copy by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/413
[Enhancement] Add get_nvcc_compiler function to retrieve nvcc path by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/414
Update lower.py to set default value for params by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/416
[Enhancement] Support Auto Layout Inference and Parallelism with variable constraint by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/417
[Enhancement] Support to find Cython path more automatically by @FrozenGene in https://github.com/tile-ai/tilelang/pull/418
[Refactor] Enhance layout inference logic in ParallelOp by @chengyupku in https://github.com/tile-ai/tilelang/pull/420
[BugFix] Fix tvm simplify pass by @smallscientist1 in https://github.com/tile-ai/tilelang/pull/421
[Enhancement] Add TMA+WS support in pipeline planning logic by @chengyupku in https://github.com/tile-ai/tilelang/pull/422
[Language] Support tile operator T.cumsum by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/423
Delete testing/python/language/test_tilelang_language_reduce_sum.py by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/424
[Bugfix] Fix a bug for simplifier by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/425
[Layout] Enhance layout inference pass by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/427
[Enhancement] Remove DeReplicate during parallel loop layout inference by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/430
[Bugfix] Fix the test data distribution of cumsum by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/432
[Enhancement] Support cute mma tile mxn8ky by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/434
[Bugfix] Removed the behavior that treated global -> local as a copy operation. by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/435
[Language] Support accumulative T.reduce_sum by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/436
[Bugfix] fix the unexpected keyword error of autotune by @yyttt6 in https://github.com/tile-ai/tilelang/pull/438
[Testing] Add atomic add test by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/439
[Typo] Rename warp_source to wrap_source by @lucifer1004 in https://github.com/tile-ai/tilelang/pull/440
[Refactor] Update KernelLaunch to clarify block name by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/441
[Enhancement] Reduce CPU overhead during kernel execution by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/437
[Enhancement] Improve layout inference accuracy in ParallelOp by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/442
[Bugfix] Fix layout inference for free fragment buffer by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/443
Bump transformers from 4.48.0 to 4.50.0 in /examples/bitnet-1.58b by @dependabot in https://github.com/tile-ai/tilelang/pull/444
[Language] Support explicit programming for identified warp groups by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/445
[Bugfix] Fix safe memory legalization for fragment store by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/446
[Refactor] Separate warp specialize rewriter and tma barrier injector pass by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/447
[Enhancement] Add new examples for warp specialization and TMA integration by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/448
[Refactor] Phaseout torch>=2.2.0 dependency by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/451
[Feature] Add TILELANG_CHECK_LAST_ERROR macro for improved error handling in CUDA and HIP by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/450
[Enhancement] Introduce pass_configs parameter for kernel Caching by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/452
[Feature] Add cache directory management functions in tilelang.cache by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/453
[Bugfix] Fix get_swizzle_layout implementation. by @cherichy in https://github.com/tile-ai/tilelang/pull/455
[Refactor] Update barrier functions and add new example for GEMM with warp specialization by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/456
[Refactor] Include examples in CI by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/457
docs: add llvm version info to installation.md. by @AsakusaRinne in https://github.com/tile-ai/tilelang/pull/459
[CI] Add elementwise and gemv examples to CI. by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/458
[Bugfix] Fix for T.copy with dynamic range by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/462
[Bugfix] Fix copy region automation for dynamic extent by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/465
[Feature] Implement fast integer power operation and related API by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/466
[Typo] Rename power_of_int with pow_of_int for consistency by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/468
[CI] Add BlocksparseGemm, Dynamic, and Cast examples to CI by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/467
[Refactor] Update set_compile_args to allow None for out_idx parameter by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/469
[Refactor] Simplify buffer_region_to_tile_region function in copy.py by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/470
[CI] Add Convolution example to CI by @xwhzz in https://github.com/tile-ai/tilelang/pull/473
[BugFix] Correct argparse for example_convolution test by @xwhzz in https://github.com/tile-ai/tilelang/pull/474
[Refactor] set USE_LLVM to optional. by @hyx1999 in https://github.com/tile-ai/tilelang/pull/476
[CI] Add Analyzer and blocksparse_attention examples to CI by @yyttt6 in https://github.com/tile-ai/tilelang/pull/472
[Refactor] Skip patchelf if not installed by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/477
[Refactor] Improve layout equality checks and error messaging by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/471
[Doc] Update version retrieval in conf.py to read from VERSION file by @xwhzz in https://github.com/tile-ai/tilelang/pull/478
Fix Device Consistency in Autotuner Threads and Add Manual Profiler Check by @yuanjypku in https://github.com/tile-ai/tilelang/pull/481
[Bugfix] Check CUDA target before checking for TMA by @gau-nernst in https://github.com/tile-ai/tilelang/pull/482
[Bugfix] Use AutoTune cache_input_tensors properly by @yyttt6 in https://github.com/tile-ai/tilelang/pull/483
Revert "[Bugfix] Use AutoTune cache_input_tensors properly" by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/488
[Enhancement] Support register input for gemm when trans_a or trans_b is true by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/490
[CI] Add flash_decoding example to CI by @xuchangtolearn in https://github.com/tile-ai/tilelang/pull/487
[CI] Add Reminder Bot for pull request contributions by @xwhzz in https://github.com/tile-ai/tilelang/pull/491
[Refactor] Introduce quantize components of TileLang and add testing for dequant gemm exmaple by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/494
[Enhancement] Introduce flag to visualize shared memory merge plan by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/496
[Refactor] Update main function structure in example scripts and add tests by @chengyupku in https://github.com/tile-ai/tilelang/pull/475
[Bugfix] Fix Hopper GEMM layout for small tile size by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/497
[Enhancement] Fallback transposed_ldmatrix into SM75_U16x4_LDSM_N when warp_n is 8 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/498
[Bugfix] Rename SM75_U16x8_LDSM_N to SM75_U16x8_LDSM_T to reflect correct matrix type by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/499
[Refactor] Update GEMM layout and operand traits for improved CUDA compatibility by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/500
[Refactor] Update JIT kernel functions and streamline GEMM tests by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/501
Fix AMD Docker issues related to conda environment setup by @Hamerlate in https://github.com/tile-ai/tilelang/pull/503
[Refactor] Refactor jit to _JitImplementation to support @tilelang.jit by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/502
[Refactor] Adjust in fragment GEMM layout by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/504
[Refactor] Update GlobalMemChecker to Detect Lower Bound illegal memory access automatically by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/505
[Enhancement] Enhance ReduceOp and JITKernel for improved dimension handling and initialization by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/507
[Refactor] Update buffer handling in layout transformation to support layout on T.view by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/509
[Bugfix] Enhance smem copy selector for uncommon shape by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/510
[Enhancement] Introduce padding annotation and improve legalize safe memory access pass by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/511
[Refactor] Enhance MergeSharedMemoryAllocations Pass for Improved Liveness Analysis and Scope Management by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/508
[Dev] Add grouped GEMM example with TileLang and PyTorch integration by @chengyupku in https://github.com/tile-ai/tilelang/pull/514
[Dev] Add grouped GEMM backward example scripts by @chengyupku in https://github.com/tile-ai/tilelang/pull/515
Fix deepgemm exmaple by @benenzhu in https://github.com/tile-ai/tilelang/pull/513
[Refactor] Support auto index bitwidth casting by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/517
[Enhancement] Support auto synchronization for global memory access by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/519
[Refactor] Replace default fp8 dtype with cute to perform fast cast by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/520
[Enhancement] Add atomicAdd for FLOAT16x2 and FLOAT16x4 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/522
[Refactor] Reorganize Thread Synchronization Steps to make sure global synchronization can be correctly lowered by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/521
[Enhancement] Add commit ID to versioning and improve logging initialization by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/524
[Enhancement] Add warp specialization attribute handling in IR and rewriter by @chengyupku in https://github.com/tile-ai/tilelang/pull/518
[CI] Add gemm and gemm_fp8 example to CI by @LeslinD in https://github.com/tile-ai/tilelang/pull/516
[Refactor] Refactor convolution example to streamline configuration and remove unused code by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/530
[Autotune] Introduce cache mechanism for auto tuner by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/527
[Refactor]: add autotune example to convolution examples by @yyttt6 in https://github.com/tile-ai/tilelang/pull/536
[Refactor] Disable legacy vectorization for buffer allocation by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/535
[Language] Support T.annotate_l2_hit_ratio via cudaStreamSetAttribute by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/539
[Bugfix] Fix a bug when simplifying warp combination for T.gemm by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/540
[AMD] Support float8 matrix core by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/537
[Doc] Include DeepWiki badge in README by @xwhzz in https://github.com/tile-ai/tilelang/pull/541
[CI] Add hadamard example to CI by @Rachmanino in https://github.com/tile-ai/tilelang/pull/549
[chore] set default build type to release if not provided by @botbw in https://github.com/tile-ai/tilelang/pull/548
[Refactor] Include several examples into ci by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/531
[AMD][Enhancement] Add support for Vectorized FP8 DataPacking by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/542
[Bugfix] Enhance layout inference pass for flexibility by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/550
[Autotune] Remove the out_idx argument from the autotune cache by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/553
[CI] Add linear attention examples to CI by @Rachmanino in https://github.com/tile-ai/tilelang/pull/552
[CI]Add norm and layout_plot by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/534

New Contributors

@FrozenGene made their first contribution in https://github.com/tile-ai/tilelang/pull/418
@lucifer1004 made their first contribution in https://github.com/tile-ai/tilelang/pull/440
@AsakusaRinne made their first contribution in https://github.com/tile-ai/tilelang/pull/459
@yuanjypku made their first contribution in https://github.com/tile-ai/tilelang/pull/481
@gau-nernst made their first contribution in https://github.com/tile-ai/tilelang/pull/482
@xuchangtolearn made their first contribution in https://github.com/tile-ai/tilelang/pull/487
@Hamerlate made their first contribution in https://github.com/tile-ai/tilelang/pull/503
@benenzhu made their first contribution in https://github.com/tile-ai/tilelang/pull/513
@Rachmanino made their first contribution in https://github.com/tile-ai/tilelang/pull/549

Full Changelog: https://github.com/tile-ai/tilelang/compare/v0.1.4...v0.1.5

Open source →

0.1.418 Apr 2025Deprecated

[Refactor] Deprecated T.Buffer as arguments and rename related calls into T.Tensor by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/281

What's Changed

[Bugfix] Support T.clear for let binding by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/268
[Bugfix] Add TMA and Producer Buffer Analysis in Warp Specialized Rewriter by @chengyupku in https://github.com/tile-ai/tilelang/pull/269
[Refactor] Improve flash attention example and layout comparison logic by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/270
[Bugfix]Add CUDA availability check in CtypesKernelAdapter by @XueSongTap in https://github.com/tile-ai/tilelang/pull/267
[CI] Add gemm performance test by @xwhzz in https://github.com/tile-ai/tilelang/pull/274
[Language] Introduce T.ptr and T.Tensor by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/276
[Refactor] Enhance Autotune by @yyttt6 in https://github.com/tile-ai/tilelang/pull/266
[Refactor] Update cache key generation in KernelCache by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/283
[Docs][Tutorial] Add tutorial for auto-tuning by @yyttt6 in https://github.com/tile-ai/tilelang/pull/285
[Refactor] Deprecated T.Buffer as arguments and rename related calls into T.Tensor by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/281
[Doc] Update README.md to correct documentation link for TileLang debug tools by @chengyupku in https://github.com/tile-ai/tilelang/pull/286
[Feature] Introduce NoSetMaxNReg for warp specialization by @chengyupku in https://github.com/tile-ai/tilelang/pull/289
[Language] Proxy tvm ir to make linter happy by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/287
[Bugfix] Enable bfloat16 atomic operations only for CUDA architectures greater than 7.5 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/291
[Doc] Update Python API docs generation by @xwhzz in https://github.com/tile-ai/tilelang/pull/278
[Doc] Remove citation page by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/292
[Dev] Correcting cxx compiler by @penguin-wwy in https://github.com/tile-ai/tilelang/pull/294
[doc/example] add gemv doc and examples by @botbw in https://github.com/tile-ai/tilelang/pull/293
[Feature] Implement ParallelLoopTransformer for enhanced loop analysis by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/295
[Enhancement] Update AtomicAdd functions for BFLOAT16 in common.h by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/297
[Refactor] Improve documentation and add detailed docstrings across multiple modules by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/298
[Bugfix] Correct method call for block reduction check when analyzing memory footprint by @NaOHCC in https://github.com/tile-ai/tilelang/pull/299
[Dynamic Symbolic] Refactor passes with dynamic symbolic and check shape bound precisely by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/302
Add autotune to conv example by @yyttt6 in https://github.com/tile-ai/tilelang/pull/301
[Bugfix] Resolve autotuner bugs for blocksparse GEMM example by @tth37 in https://github.com/tile-ai/tilelang/pull/300
[Bugfix] Replace profiler.mod with profiler.adapter to fix AttributeError by @LeslinD in https://github.com/tile-ai/tilelang/pull/305
[Enhancement] Add support for CUDA architecture 8.9 in GEMM template by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/304
[BugFix] Fix unintended Git config overrides in CI runners by @xwhzz in https://github.com/tile-ai/tilelang/pull/306
[Cache] Implement in-memory cache by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/308
[Bugfix] Updated autotune usage in the examples to align with the latest changes by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/309
[Bugfix] Fix dynamic axis with variable extent by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/311
[Bugfix] Fix layout conflict issue for gqa decoding examples by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/314
[Bugfix] Fixed the handling logic of IfThenElseNode in if_stmt_binding by @chengyupku in https://github.com/tile-ai/tilelang/pull/315
[Bugfix] Fix logic error in ReduceOp when handling CUDA architecture by @chengyupku in https://github.com/tile-ai/tilelang/pull/316
[CostModel] Introduce cuda driver api to get precise shared memory capacity by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/317
[Dev] Add FP8 Quantization Examples and Absolute Maximum Reduction Operation Support by @chengyupku in https://github.com/tile-ai/tilelang/pull/320
[Tools] Summarize TFLOPS Information from a tilelang program by @yyttt6 in https://github.com/tile-ai/tilelang/pull/321
Support block_N sizes that are 2^n in deepgemm example by @zcnrex in https://github.com/tile-ai/tilelang/pull/319
[Feat] Enhance CUDA Property Handling by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/322
[Bugfix] add a patch to fix T.abs on float16 by @botbw in https://github.com/tile-ai/tilelang/pull/325
[AMD] Adapt rocm and support T.gemm with transpose_b=False for amd backend by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/327
[Dynamic Symbolic] Adaptively vectorize with different condition expressions by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/326
[Bugfix] Fix fragment layout annotation in example gqa decode by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/329
[AMD] Support Transpose_A=True and GEMM_RS for hip backend by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/331
[Refactor] Optimize RMS normalization kernel in rms_norm.py by @chengyupku in https://github.com/tile-ai/tilelang/pull/333
[AMD] Fix for missing composable kernel include path when compile kernels on amd gpus by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/334
[Example] Add sparse gqa decode example by @xiayuqing0622 in https://github.com/tile-ai/tilelang/pull/332
[Enhancement] Enhance FP8/FP4 type handling in CUDA codegen by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/323
[Doc] Fix typo and heading level in GEMV tutorial by @yeh-sudo in https://github.com/tile-ai/tilelang/pull/337
[Dev] Add Group Cast FP8 Example by @chengyupku in https://github.com/tile-ai/tilelang/pull/338
[Enhancement] Support region padding when convert buffer load to buffer region by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/342
[Example] Add triton block sparse gqa decode by @YizhaoGao in https://github.com/tile-ai/tilelang/pull/341
[Enhancement] Support index bit width configuration by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/343
[Bugfix] Fix X_amax Correctness Issue in Group Cast FP8 by @chengyupku in https://github.com/tile-ai/tilelang/pull/345
[Bugfix] Fix Transposed Fragment Layout for amd GEMM_RS matrix core by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/346
[AutoTune] Refactor AutoTuneArtifact to utilize kernel as context instead of profiler by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/344
[Bugfix] Compile/"cached" still not loading cached kernel for example in example_mha_bwd by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/339
[Refactor] Implement thread-local storage for FrameStack in frame.py and kernel.py by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/352
[Typo] Replace kernel.func with kernel in mla benchmark scripts by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/354
[AMD][Docker] Create Dockerfile for ROCm environment setup by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/355
[Enhancement] Update group_per_split_token_cast_to_fp8 to support multiple data types by @chengyupku in https://github.com/tile-ai/tilelang/pull/356
[Enhancement] Support pass config disable_warp_specialize to disable auto specialization on hopper by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/357
[Example] Introduce autotuning example for GEMM with enhanced configuration options by @chengyupku in https://github.com/tile-ai/tilelang/pull/360
[Example] Handle Scenarios in Which a Threadblock is Assigned Only Invalid Block Indices for Sparse Attention by @xiayuqing0622 in https://github.com/tile-ai/tilelang/pull/361
[Bugfix] Correct dynamic shared memory size error handling in HIP by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/362
[AMD] Implement Deepseek MLA for AMD by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/363
[Bugfix] Fix compilation issues for amd cdna element size check by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/364
[AMD] Support FlashMLA with num split template for AMD gpus by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/366
[MLA][AMD] Add amd mla benchmarking by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/367
[Bugfix] Adjust Autotuner threadpool max_workers limit to available CPUs by @tth37 in https://github.com/tile-ai/tilelang/pull/368
[Language] Introduce T.any_of and T.all_of to reduce a bool arrary by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/371
[AMD][Setup] Support HIP in setup.py by @zhhangBian in https://github.com/tile-ai/tilelang/pull/369
[Typo] Remove debug print by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/373
[Docs] Add AMD Flash MLA Documentation to Tutorials Section by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/376
[Bugfix] Add filelock for cython build by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/377
[Typo] Remove unused comments generated by copilot by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/379
[Doc] Add deepseek_mla to documentation index by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/380
[Refactor] Remove debug message in pass legalize_safe_memory_access by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/381
[Enhancement][Pipeline] More precise copy code block detection in pipeline by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/384
[Revert] Revert modifications for pass FlattenBuffer by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/385
[Dynamic Symbolic] Add pass_config to customize vectorization and tail split by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/383
[Pytest Fix] Wrap tests in dynamic benchmark by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/387
[Doc] Update README.md for deepseek_mla on AMD by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/389
[Pipeline][Enhancement] Add copy_prepare stage to support mask and index caching by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/392
[Refactor] Refactor warp_specialized_rewriter to support multiple acquire/release patterns by @chengyupku in https://github.com/tile-ai/tilelang/pull/391
[Enhancement] Report Error Body in ParallelOp Layout Inference by @chengyupku in https://github.com/tile-ai/tilelang/pull/394
[Bugfix] Support T.Parallel with local register assignment by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/395
[Enhancement] Introduce a smarter warp partition strategy by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/396
[Example] Add bitnet-1.58b examples by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/399
Bump transformers from 4.40 to 4.48.0 in /examples/bitnet-1.58b by @dependabot in https://github.com/tile-ai/tilelang/pull/400
[BugFix] Address should aligned with access size in tail split by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/401
[Enhancement] Move T.any_of and T.all_of op registration from python into cpp by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/398
Add preliminary support for bf16 for AMD by @OscarSavolainen in https://github.com/tile-ai/tilelang/pull/388
[BugFix] Conditions Robustness in dynamic vectorize by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/404
[CI] Update CI configuration to run pytest with automatic parallelization by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/393
[Documentation] Fix Installation Documentation by @andyluo03 in https://github.com/tile-ai/tilelang/pull/405

New Contributors

@XueSongTap made their first contribution in https://github.com/tile-ai/tilelang/pull/267
@botbw made their first contribution in https://github.com/tile-ai/tilelang/pull/293
@NaOHCC made their first contribution in https://github.com/tile-ai/tilelang/pull/299
@tth37 made their first contribution in https://github.com/tile-ai/tilelang/pull/300
@LeslinD made their first contribution in https://github.com/tile-ai/tilelang/pull/305
@zcnrex made their first contribution in https://github.com/tile-ai/tilelang/pull/319
@xiayuqing0622 made their first contribution in https://github.com/tile-ai/tilelang/pull/332
@yeh-sudo made their first contribution in https://github.com/tile-ai/tilelang/pull/337
@YizhaoGao made their first contribution in https://github.com/tile-ai/tilelang/pull/341
@zhhangBian made their first contribution in https://github.com/tile-ai/tilelang/pull/369
@dependabot made their first contribution in https://github.com/tile-ai/tilelang/pull/400
@OscarSavolainen made their first contribution in https://github.com/tile-ai/tilelang/pull/388
@andyluo03 made their first contribution in https://github.com/tile-ai/tilelang/pull/405

Full Changelog: https://github.com/tile-ai/tilelang/compare/v0.1.3...v0.1.4

Open source →

0.1.323 Mar 2025

[Docker] Add libstdcxx-ng-12 to Dockerfiles for CUDA versions by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/160

What's Changed

[Docker] Add libstdcxx-ng-12 to Dockerfiles for CUDA versions by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/160
Add cpu jit with backend ctypes by @xs-keju in https://github.com/tile-ai/tilelang/pull/154
[Carver] Multi-Threads Compilation for Fast Auto Tuning by @SiriusNEO in https://github.com/tile-ai/tilelang/pull/156
[Refactor] Replace T.If with native Python if statement for mla paged kernel by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/162
[Enhancement] Improve CUDA path detection by @xwhzz in https://github.com/tile-ai/tilelang/pull/157
[Refactor] Replace T.thread_binding with T.get_thread_binding in examples and test cases by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/163
[Bugfix] Cast bool dtype into int8 in blocksparse examples by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/167
[Example] Implement NSA Decode tilelang exampls by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/168
[Release] Bump version to v0.1.2.post1 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/166
Use SS-GEMM for PV in mla by @YouJiacheng in https://github.com/tile-ai/tilelang/pull/165
[Example] Implement tilelang native sparse attention varlen example by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/170
[Bugfix] Implement boundary check for the buffer shape with dynamic symbolic by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/173
[AutoTune] Enable config-performance trace by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/174
[Feat] Append Pass Context and TMA lowering configuration option by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/175
[Feat] Introduce new caching mechanism for compiled kernels by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/176
[Refactor] Enhance GPU Kernel Launch with Environment Thread Creation by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/178
[Bugfix] Improve Thread Variable Handling in Layout Inference by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/179
[Examples] Implement NSA Backward kernels by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/180
[Enhancement] Optimize CMake build process with dynamic job count calculation by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/183
[Bugfix] Add dynamic shape support with out_idx in Cython JIT kernel compilation by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/185
[Dev][Bugfix] Add RMS Normalization Kernels and Fix Reduce Bug by @chengyupku in https://github.com/tile-ai/tilelang/pull/188
[Dev] Add the failed nvcc command to the exception message by @penguin-wwy in https://github.com/tile-ai/tilelang/pull/189
[Bugfix] Fix T.copy for scalar datatypes by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/190
[Enhancement] Simplify GEMM example with direct kernel compilation by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/191
[Bugfix] Make quickstart work properly on cu118 by @penguin-wwy in https://github.com/tile-ai/tilelang/pull/193
[Language] Support clamp in language by @hyx1999 in https://github.com/tile-ai/tilelang/pull/192
[Refactor] Add SetMaxNRegCollector to Improve Register Hint Handling in Warp Specialized Rewriter by @chengyupku in https://github.com/tile-ai/tilelang/pull/194
[Feature] Add TMA Store Synchronization Support by @chengyupku in https://github.com/tile-ai/tilelang/pull/195
Update expired example code. by @66RING in https://github.com/tile-ai/tilelang/pull/196
[CMake] Add CUDA Major Version Detection for Conditional Compilation by @chengyupku in https://github.com/tile-ai/tilelang/pull/197
[Feature] Support Async Pipeline inference within if scope by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/198
[Dev] Add new example for FlashAttention with pipelined execution by @chengyupku in https://github.com/tile-ai/tilelang/pull/200
[Enhancement] Enhancing the handling of conditional statements in the pipeline by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/201
[Feature] Upgrade cutlass version and support fp8 T.gemm by @zqh-wz in https://github.com/tile-ai/tilelang/pull/202
[Docker] Update Dockerfiles to specify exact version of libstdcxx-ng by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/203
[Dev] Add GQA backward example by @chengyupku in https://github.com/tile-ai/tilelang/pull/205
[LICENSE] Typo fix in LICENSE by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/208
[Enhancement] Allow mma fallback when wgmma is not supported by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/206
[Examples] Expand tuning configurations for FlashAttention example by @chenghuaWang in https://github.com/tile-ai/tilelang/pull/204
[Enhancement] Avoid tvm ffi handling when out_idx is specified by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/209
[Fix] Fix K // block_K to T.ceildiv(K,block_K) and add tests by @hyx1999 in https://github.com/tile-ai/tilelang/pull/210
[Dev] Implement IfStmtBinding and MergeIfStmt transformations by @chengyupku in https://github.com/tile-ai/tilelang/pull/211
[Language] Introduce T.reshape and T.view by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/212
[Enhancement] Improve device handling in Cython kernel adapter by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/220
[Enhancement] Update format script to support force compare with upstream by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/221
[Refactor] Introduce KernelParam integration across modules by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/223
[Bugfix] Fix mismatch of shared memory layout and mma atom on Hopper by @zqh-wz in https://github.com/tile-ai/tilelang/pull/224
[Refactor] Update kernel compilation and profiling in examples by @chengyupku in https://github.com/tile-ai/tilelang/pull/225
[Examples] Add fp8 gemm 2xAcc and deepgemm example by @cherichy in https://github.com/tile-ai/tilelang/pull/217
[Doc] Add instructions for installing nightly version by @xwhzz in https://github.com/tile-ai/tilelang/pull/226
[Bugfix] Disable force inline for ldmatrix by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/227
[Bugfix] Support duplicate tma desc declaration by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/228
[Refactor] Rename clamp functions and enhance dtype handling in tests by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/232
[Enhancement] Simplify kernel source extraction in JIT adapters by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/230
[Feature] Add reduce_max corresponding tests by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/236
[BugFix] Fix bug of missing MBarrierExpectTX by @chengyupku in https://github.com/tile-ai/tilelang/pull/241
[Refactor] Refactor for Better Layout Conflict Handling by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/240
[Refactor] Align torch_assert_close tensor comparison with torch.testing.assert_close by @xwhzz in https://github.com/tile-ai/tilelang/pull/239
[Dev] Implement FlashAttention3 Backward by @chengyupku in https://github.com/tile-ai/tilelang/pull/244
[BugFix] Fix bug of mismatching dtype in testing by @xwhzz in https://github.com/tile-ai/tilelang/pull/245
[Enhancement] Add zero initialization option to GEMM operations by @chengyupku in https://github.com/tile-ai/tilelang/pull/246
[Enhancement][CUDA] Avoid C7508 for CUDA backend via assigning default value to minBlocksPerMultiprocesor by @cherichy in https://github.com/tile-ai/tilelang/pull/248
[Feature] Add database storage for JITKernel cache with Cython and Ctypes adapters by @Alex4210987 in https://github.com/tile-ai/tilelang/pull/213
[Examples] Implement elementwise add kernel by @chenghuaWang in https://github.com/tile-ai/tilelang/pull/219
[Refactor] Phaseout LLVM Dependency by Making it Optional by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/247
[Readme] Update Bib Citation Section by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/249
[Enhancement] Support float variable as arguments by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/250
add autotune to example_gemm.py by @yyttt6 in https://github.com/tile-ai/tilelang/pull/252
[Language] Introduce T.alloc_var to define a variable like int var; by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/255
[Example] Implement Kernel Example cumsum by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/258
[Refactor] Refactor CUDA post-processing callback registration in TileLang by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/259
[Refactor] Move compilation outside critical section by @YouJiacheng in https://github.com/tile-ai/tilelang/pull/260
[CI] Use auditwheel to generate manylinux wheels by @oraluben in https://github.com/tile-ai/tilelang/pull/251
[Bugfix] Fix Benchmark/Example Code for Autotuning by @SiriusNEO in https://github.com/tile-ai/tilelang/pull/254
[Language] Enhance alias to support blockwise memory load by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/261
[Bugfix] Fix auto tuning tma handling by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/263
[Release] Bump version to 0.1.3 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/264

New Contributors

@xs-keju made their first contribution in https://github.com/tile-ai/tilelang/pull/154
@YouJiacheng made their first contribution in https://github.com/tile-ai/tilelang/pull/165
@penguin-wwy made their first contribution in https://github.com/tile-ai/tilelang/pull/189
@hyx1999 made their first contribution in https://github.com/tile-ai/tilelang/pull/192
@66RING made their first contribution in https://github.com/tile-ai/tilelang/pull/196
@zqh-wz made their first contribution in https://github.com/tile-ai/tilelang/pull/202
@chenghuaWang made their first contribution in https://github.com/tile-ai/tilelang/pull/204
@cherichy made their first contribution in https://github.com/tile-ai/tilelang/pull/217
@Alex4210987 made their first contribution in https://github.com/tile-ai/tilelang/pull/213
@yyttt6 made their first contribution in https://github.com/tile-ai/tilelang/pull/252
@oraluben made their first contribution in https://github.com/tile-ai/tilelang/pull/251

Full Changelog: https://github.com/tile-ai/tilelang/compare/v0.1.2...v0.1.3

Open source →

0.1.012 Feb 2025

[LICENSE] Add LICENSE for flashinfer by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/19

What's Changed

[LICENSE] Add LICENSE for flashinfer by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/19
[Doc] Fix installation scripts and docs for dequantize gemm by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/20
[Doc] Use sphinx to generate docs. by @xwhzz in https://github.com/tile-ai/tilelang/pull/21
[Doc] update installation.md and readme by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/22
[Doc] fix a typo in installation.rst by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/24
[Doc] Remove legacy files and update reference by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/25
[CI][Test] Add test cases for tilelang transform AnnotateDeviceRegions and MakePackedAPI by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/26
[Doc] Create a workflow to host docs using GitHub Pages. by @xwhzz in https://github.com/tile-ai/tilelang/pull/28
[CI][Test] Add test cases for tilelang transform InjectSoftwarePipeline and FrontendLegalize by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/30
[Bugfix] Replace thread binding detector in LayoutInference Pass by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/31
[CI] Comprehensive Test cases Implementation of Matmul Dequantize by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/32
[Doc] Update GitHub Actions workflow for documentation deployment and add CNAME file. by @xwhzz in https://github.com/tile-ai/tilelang/pull/33
[Refactor] Simplify interface via replacing argument thread binding of intrinsics with KernelFrame.Current by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/34
[Bugfix] Reorder Passes: Place Vectorize Loop Before StorageFlatten and FlattenBuffer to Prevent Redundant Allocations by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/37
[Doc] Update documentation structure and content by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/39
[Doc][CI] Update GitHub Actions workflow for documentation build and deployment. by @xwhzz in https://github.com/tile-ai/tilelang/pull/42
[CI] Allow manual triggering of documentation workflow in addition to… by @xwhzz in https://github.com/tile-ai/tilelang/pull/43
[CI][Test] Add test cases for tilelang transform PipelinePlanning by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/44
[CI][Test] Add test cases for tilelang transform LayoutInference and LowerTileOp on loop tail split functionality by @tzj-fxz in https://github.com/tile-ai/tilelang/pull/29
[Debug] Introduce T.print for buffer and variables logging on frontend by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/45
[CI] Change pull request trigger to pull_request_target for documen… by @xwhzz in https://github.com/tile-ai/tilelang/pull/48
[Dev] Add FlashDecoding example by @chengyupku in https://github.com/tile-ai/tilelang/pull/46
[Doc] update README that tilelang has been used in AttentionEngine by @smallscientist1 in https://github.com/tile-ai/tilelang/pull/50
[Doc] Remove unnecessary layout annotation by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/49
[CI][Test] Add test cases for tilelang kernel convolution by @chengyupku in https://github.com/tile-ai/tilelang/pull/51
[Dev] Implement test case for tilelang transformations by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/53
[CI][Test] Add test cases for tilelang kernel FlashAttention by @chengyupku in https://github.com/tile-ai/tilelang/pull/54
[CI][Test] Add test cases for element_add by @Cunxiao2002 in https://github.com/tile-ai/tilelang/pull/47
[CI] Clean up target repository before publishing documentation. by @xwhzz in https://github.com/tile-ai/tilelang/pull/55
[CI][Test] Add test cases for tilelang transform ClusterPlanning by @chengyupku in https://github.com/tile-ai/tilelang/pull/57
[Doc] Append debug relevant testing and documentations by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/58
[CI][Test] Add test cases for tilelang transform LowerHopperIntrin by @chengyupku in https://github.com/tile-ai/tilelang/pull/59
[Doc] Add matmul kernel tutorial with tile library by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/60
[Dev] Separate LoopVectorize Pass from upstream tvm by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/62
[Dev] Support FP8 Codegen for cuda backend by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/64
[Dev] Add test case for bfloat16 and int4 gemm with mma by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/65
[CI][Test] Add test cases for tilelang transform InjectFenceProxy by @chengyupku in https://github.com/tile-ai/tilelang/pull/66
[Tools] Introduce plot_layout to visualize the fragment layout by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/68
[Dev] Remove unnecessary python dependencies by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/69
[Carver] Introduce a tile-structure based cost model for auto tuning by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/70
[Bugfix] bug fix for bitblas dependency by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/71
[CI][Test] Add test cases for tilelang transform MultiVersionBuffer and WarpSpecialized by @chengyupku in https://github.com/tile-ai/tilelang/pull/72
[CostModel][Carver] Support Hint Recommend for Shared memory Kernel Fusion by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/73
[Carver] Remove legacy todo items in carver's readme by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/74
[Dev] Add mha backward example by @chengyupku in https://github.com/tile-ai/tilelang/pull/77
[Release] Bump version into v0.1.0 by @LeiWang1999 in https://github.com/tile-ai/tilelang/pull/76

New Contributors

@xwhzz made their first contribution in https://github.com/tile-ai/tilelang/pull/21
@Cunxiao2002 made their first contribution in https://github.com/tile-ai/tilelang/pull/22
@tzj-fxz made their first contribution in https://github.com/tile-ai/tilelang/pull/29
@chengyupku made their first contribution in https://github.com/tile-ai/tilelang/pull/46
@smallscientist1 made their first contribution in https://github.com/tile-ai/tilelang/pull/50

Full Changelog: https://github.com/tile-ai/tilelang/compare/v0.0.1...v0.1.0

Open source →

tilelang

Release timeline

Releases

What's Changed

New Contributors

What's Changed

New Contributors

What's Changed

New Contributors

What's Changed

New Contributors

What's Changed

New Contributors

Why we need this post release?

What's Changed

New Contributors

What's Changed

New Contributors

What's Changed

What's Changed

New Contributors

What's Changed

New Contributors