[Release] v0.24.0 Release Candidate Notes

[ysh329]

Introduction

The TVM community has worked since the last release to deliver the following new exciting improvements!

The main tags are below (bold text is with lots of progress): Relax etc.

Please visit the full listing of commits for a complete view: v0.24.dev0...v0.24.0.rc0.

Community

None.

RFCs

None.

Adreno

• #18867 - Revive and consolicate Adreno features

Arith

• #19417 - Expose allow_override parameter in Python Analyzer.bind()

BugFix

• #19432 - [Fix][CUDA] Version compatibility of CUDA symbols
• #19427 - [FIX] Skip metal target tag registration for unsupported LLVM CPUs
• #19390 - [LLVM] Fix insertDeclare API mismatch for ROCm-bundled LLVM 20
• #19410 - [Fix][Runtime][RPC] Fix remote tensor handle cleanup for RPC return values
• #19385 - [MetaSchedule] Fix compile_relax to apply MetaScheduleApplyDatabase after FuseOps
• #19383 - [TIRx] Fix bad-optional-access in BF16/FP8 legalize passes for target-less PrimFuncs
• #19382 - [TIRx] Fix VerifyMemory crash for PrimFuncs without target attribute
• #19380 - [TOPI] Fix get_const_tuple hanging indefinitely when passed a te.Tensor
• #19368 - Align tir.round to ties-to-even across all backends
• #19367 - [ONNX] Fix Round op to use ties-to-even
• #19362 - [TVMScript] Fix invalid f-string format spec causing TypeError on Python 3.14
• #19352 - [TVMScript] Add doc.keyword handling for ExprEvaluator._visit
• #18957 - [FIX] Inline ceil_log2 in gpu_2d_continuous_cumsum to fix MakePackedAPI error
• #18940 - [Fix] Fix tvm.tir references in Tflite frontend
• #18887 - [FIX] Fix cumsum kernel sblock_alloc_buffer for non-sblock buffer
• #18881 - [FIX][Adreno] Replace AllocBuffer with Bind in texture alloc injection
• #18838 - [TOPI] Fix resize accuracy issue with non-floor rounding
• #18782 - [S-TIR][FIX] Remove redundant std::move() to itself
• #18742 - [Fix] Handle empty variable name in NameSupply::FreshName
• #18694 - [TIR] Fix incorrect optimization when lowering floordiv and f…
• #18695 - [FIX] Fix T.sblock due to concurrent merge

CI

• #19445 - [REFACTOR] Decouple data.py from Jenkins script and docker images
• #18827 - Update images to 20260301-134651-63f099ad
• #18863 - [S-TIR][Test] Mark meta_schedule tuning tests as skip
• #18851 - Remove stale test scripts (i386, hexagon, mypy)
• #18850 - [TEST] Remove stale URL mappings from request_hook
• #18848 - Remove legacy lint scripts and Apache RAT
• #18817 - [REFACTOR]Further cleanup docker images
• #18812 - [REFACTOR]Modernize Python dependency management with uv
• #18809 - Add GitHub Actions lint workflow
• #18805 - [REFACTOR][TEST] Migrate tir-transform tests from TE to TVMScript
• #18804 - [REFACTOR][TEST] Remove unused te imports from test files
• #18800 - Update images to 20260219-160550-72f51851
• #18796 - Refactor Dockerfiles and installation scripts
• #18775 - Update images to 20260214-152058-2a448ce4
• #18783 - Update system cuda version 12.4->12.8
• #18780 - Remove unity from tvm-bot
• #18777 - Update Pillow, pytest-rerunfailures, junitparser, xgboost, onnx and pytorch
• #18647 - Upgrade Python to 3.10 in CI
• #18749 - Remove i386 and Hexagon from CI pipeline (2)
• #18757 - Further cleanup CI after merging unity to main test
• #18456 - Move conda config files to tests/conda and remove unused conda build infrastructure
• #18755 - [TEST] Cleanup legacy tests and migrate unity tests to main one
• #18737 - Remove i386 and Hexagon from CI pipeline (1)
• #18748 - Remove i386 and hexagon from .asf.yaml
• #18719 - [REFACTOR][TEST] Migrate all codegen test to tvmscript
• #18717 - Fix double newlines in nightly docker update
• #18711 - [REFACTOR][TEST] Replace CompareBeforeAfter for pytest compact
• #18692 - Fix NameError in nightly docker update workflow

Docker

• #18854 - Refactor bash.sh: auto-detect rootless, add --shell, TVM_DEV_MOUNTS
• #18710 - [ci]Nightly Docker image update

Docs

• #19439 - Refactor BYOC example NPU tutorial
• #19414 - Fix stale tvm.tirx exclude list and add missing legalize_ops.unary entry
• #19409 - Fix outdated source install and API reference docs
• #19407 - Fix [Docs] python -c "import tvm; print(tvm.file)" fail #18714: python -c "import tvm; print(tvm.file)" fail
• #19396 - Add code generation architecture documentation
• #19398 - Add TVMScript architecture documentation
• #19397 - Add PyModule tutorial to How-To toctree
• #19399 - Clean up architecture docs: remove duplicates, fix stale content
• #19389 - Add Relax VM architecture documentation
• #19394 - Add operator fusion architecture documentation
• #19395 - Add BYOC external library dispatch architecture documentation
• #19387 - Add docstrings for nn.Module classes and core APIs in relax.frontend.nn
• #19386 - Add tvm.s_tir.tensor_intrin API reference and remove empty legacy tvm/tir directory
• #19379 - Add API reference for tvm.arith, tvm.testing, tvm.exec, tvm.tirx.backend and extend topi/contrib/ir/target docs
• #19369 - Add API reference for tvm.s_tir submodules: dlight, meta_schedule, backend
• #19366 - Add API reference documentation for tvm.script module
• #19356 - Add DLight and MetaSchedule deep-dive instructions
• #19364 - TFLite tests requiring Python 3.10 and specific package versions to avoid core dumps
• #19354 - Add tutorial for importing models from PyTorch, ONNX, and TFLite
• #19358 - Add Dataflow Pattern Language (DPL) documentation for Relax
• #19357 - Add Disco distributed runtime architecture overview
• #19351 - Fix outdated paths, links, and add missing API references across documentation(3)
• #19353 - Add tvm.s_tir.analysis API reference page
• #19350 - Add Relax VM architecture overview in documentation
• #19344 - Fix outdated code examples, typos, and missing API reference in documentation(2)
• #18965 - Fix outdated code examples, types, and missing references across documentation
• #18966 - [DOC] Fix various issues
• #18953 - Align documentation with tirx/s_tir namespace split
• #18947 - Add tutorial for mixing Python/PyTorch with TVM using BasePyModule
• #18939 - [DOC] Fix inconsistent code comments
• #18941 - Fix duplicate license headers and incorrect module paths after tirx rename
• #18908 - Clean up stale references from recent refactors
• #18906 - Update outdated references from recent refactors
• #18860 - [CI]Update Sphinx dependencies
• #18855 - Fix RPC tutorial to use set_input + invoke_stateful API
• #18808 - [DOC] Update installation docs with missing dependencies ([Bug] the latest doc about installing tvm from source code is not complete. #18194)
• #18799 - [DOC] Fix docstring, unify CMake, nvidia-docker deprecation
• #18797 - [DOC] Unify CUDA naming
• #18794 - [DOC] Unify GitHub naming
• #18770 - [DOC] Fix PYTHONPATH in "Install from Source"
• #18732 - [DOC] Fix RST syntax
• #18753 - [DOC] Fix the loop length in a loop tiling example
• #18731 - [DOC] Fix grammar
• #18718 - Clarify trusted usage

Frontend

• #19401 - [TFLite] Add test coverage for SHAPE and RANGE operators
• #19402 - [Test][TFLite] Add unit tests for PRELU
• #19400 - [TFLite] Add TILE operator tests and edge cases
• #19388 - [Test][TFLite] Add unit tests for LEAKY_RELU, HARD_SWISH ReLU_N1_to_1 and LOG_SOFTMAX
• #19365 - [Test][TFLite] Add unit tests for RESIZE_BILINEAR and RESIZE_NEAREST_NEIGHBOR ops
• #18970 - [TFLite]Add expected IRModule checks for conv2d, pool2d, and batch_matmul tests
• #19341 - [ONNX] Fix SplitToSequence keepdims=0 and uneven last chunk
• #18969 - [ONNX] Support select_last_index for ArgMax and ArgMin
• #18951 - [ONNX] Add MatMulInteger support to Relax ONNX frontend
• #18946 - [ONNX] Add If operator support to Relax ONNX frontend
• #18929 - [TFLite] Fix undefined symbols and Relay API remnants in TFLite frontend
• #18773 - [ONNX] Handle Gelu approximate attribute from Opset 20

LLVM

• #18909 - [Target]Fix -mcpu validation compatibility across LLVM versions
• #18853 - Bump minimum LLVM version to 15
• #18818 - Fix build failures when building with llvm>=22
• #18772 - [Codegen] Cast NaN to bool gives true
• #18706 - Fix insertDbgValueIntrinsic for Metal backend

MetaSchedule

• #19438 - [S-TIR]Make evolutionary search resilient to trace replay failures

Metal

• #19493 - Include logging headers for metal
• #18877 - Batched command dispatch and staging buffer pool
• #18819 - [REFACTOR]Update CHECK_LE to TVM_FFI_ICHECK_LE in Metal runtime
• #18811 - [Refactor]Update ICHECK to TVM_FFI_ICHECK in Metal runtime

ROCm

• #15518 - Fix some ROCm codegen bugs

Relax

• #19492 - [BugFix]Add legalize for isnan, isinf, isfinite
• #19489 - [Frontend][TFLite] Add BROADCAST_TO, EMBEDDING_LOOKUP, and SELECT_V2
• #19490 - [Frontend][TFLite] Add SCATTER_ND operator for Relax TFLite
• #19467 - [ONNX] Fix CumSum axis handling: support runtime axis tensor, error on multi-element axis
• #19473 - [Frontend][TFLite] Add RANDOM_UNIFORM, RANDOM_STANDARD_NORMAL, and MULTINOMIAL
• #19487 - [Frontend][TFLite] Add BROADCAST_ARGS operator mapping
• #19481 - [Frontend][TFLite] Add DILATE operator mapping
• #19485 - [Frontend][TFLite] Add ATAN2 op and TFLite mapping
• #19480 - [BugFix][ONNX] Fix ConstantOfShape converter when value attr is absent
• #19468 - [Frontend][TFLite] Fix STRIDED_SLICE negative stride and add STRIDED_SLICE/SPLIT_V tests
• #19421 - [Frontend][TFLite] Add DENSIFY operator test and fix prefetched handling
• #19464 - [Frontend][TFLite] Add NON_MAX_SUPPRESSION_V4 converter
• #19466 - [Frontend][TFLite] Add BITCAST operator mapping
• #19433 - [Frontend][TFLite] Fix dynamic FILL/SPLIT_V partial implementations
• #19426 - [Frontend][TFLite] Add soft-NMS support for TFLite NON_MAX_SUPPRESSION_V5
• #19450 - [BugFix][ONNX] Honor auto_pad in ConvTranspose converter
• #19431 - [Frontend][KVCache] Extend masked sequence prefill to causal left-padding
• #19434 - [Frontend][TFLite] Add CUMSUM operator mapping
• #19430 - [NN] Use int64 for RoPE apply flag
• #19428 - [FRONTEND][ONNX] Support Softmax, LogSoftmax and Hardmax when opset version ≤12
• #19425 - [Backend]Add NPU BYOC backend example
• #19424 - Fix deprecation warning
• #19416 - [TVMScript] Print ExternFunc struct_info when non-default
• #19415 - [Frontend][TFLite] Fix bool REDUCE_ANY/REDUCE_ALL compile failure
• #19413 - [Frontend][TFLite] Add REDUCE_ANY and REDUCE_ALL
• #19411 - fix 'occured' -> 'occurred' in transform.h doc comment
• #19408 - [Frontend][TFLite] Fix and test MATRIX_DIAG, MATRIX_SET_DIAG, SPARSE_TO_DENSE
• #19405 - [Frontend][KVCache] Restructure kv_cache kernels
• #19404 - [tflite] Add PRELU/LRN/SQUARED_DIFFERENCE tests (partial [Tracking Issue][TFLite] Expand unit test coverage for supported non-quantized operators #18971)
• #19392 - [Frontend][KVCache] Add masked sequence prefill helper for encoder valid lengths
• #19372 - [frontend][tflite] Add tests for fully_connected/depthwise_conv2d/transpose_conv/l2_pool2d
• #19391 - [ONNX] Add frontend support for QuantizeLinear, DequantizeLinear, and DynamicQuantizeLinear
• #19384 - [BugFix]Select target-specific pipeline in tvm.compile when GPU target is provided
• #19371 - [frontend][tflite] Add tests for l2_normalization/slice/reverse_v2
• #19345 - [Frontend][TFLite] Implement DETECTION_POSTPROCESS tflite operator
• #19381 - [TFLite] Fix and test DEPTH_TO_SPACE/SPACE_TO_DEPTH, SELECT ops
• #19373 - [TFLite] Fix MIRROR_PAD/ONE_HOT converters and add tests for PAD, PADV2, MIRROR_PAD, TOPK_V2, ONE_HOT
• #19370 - [TFLite] Add test coverage for Reduction operations ([Tracking Issue][TFLite] Expand unit test coverage for supported non-quantized operators #18971)
• #19361 - [ONNX] Support ConcatFromSequenc/SequenceInsert with new_axis=1
• #19349 - [TFLite] Add NON_MAX_SUPPRESSION_V5 support
• #18963 - [ONNX] Support Resize dynamic ROI via TOPI
• #18955 - [ONNX] Fix shape/dynamic restrictions for Squeeze/Unsqueeze and Slice
• #18956 - [ONNX] Complete ShapeExpr reshape handling in ONNX frontend
• #18950 - [ONNX] Add Optional and MatMulInteger16 frontend support
• #18952 - [ONNX] Add roi_pool op and MaxRoiPool frontend support
• #18948 - Add conv3d_transpose and ONNX ConvTranspose 3D support
• #18943 - [Vision] Add get_valid_counts and classic NMS
• #18942 - [TOPI] Add relax.vision.multibox_transform_loc for SSD/TFLite box decode
• #18933 - Add affine_grid operator with PyTorch and ONNX frontend support
• #18937 - [PyTorch] Add 3D interpolate support using resize3d
• #18936 - [ONNX][Torch] Add roi_align support and frontend integration
• #18931 - [ONNX] Add image.resize3d op and wire 5D Resize
• #18932 - [ONNX] Add GridSample ONNX frontend integration
• #18868 - [TFLite] Introduce TensorFlow Lite frontend
• #18869 - [LAYOUT] Support multiple axis paching
• #18904 - [PyTorch] Add torch.cond support to ExportedProgram frontend
• #18870 - Add input type validation for make_shape and corresponding tests
• #18903 - [PyTorch] Fix crash on dynamic shapes with identity slice in ExportedProgram importer
• #18878 - [ONNX] Support dynamic repeats for Tile
• #18864 - [Refactor] Phase out FewShotTuning
• #18814 - Make ShapeType ndim parameter mandatory
• #18815 - [PyTroch] Add randn.default and randn_like.default support
• #18520 - Fix llama4_rope_with_position_map to support partial rotary factor
• #18764 - Add size heuristic to skip folding large creation ops
• #18762 - Remove TODO comment for moving code in fuse_tir.cc
• #18733 - Migrate NN conv/pooling/grad attrs from Array to Array<int64_t>
• #18736 - Support constant folding for call_tir with tuple outputs
• #18726 - [PyTorch] Simplify tensor args conversion in Dynamo
• #18725 - [PyTorch] Fix scalar parameter inputs in Dynamo
• #18670 - [Torch] Avoid decomposition crash with sparse CSR buffers
• #18704 - [Onnx][BatchNorm] Pass momentum and training_mode into BatchNorm Operator
• #18661 - [Python]Fix YaRN correction dim calculation
• #18691 - [Onnx][Resize] Fix ROI values when tensor ROI is Empty pass node Constant
• #18673 - [Onnx] Support Multi Input Ops with Multidirectional Broadcasting
• #18677 - [NN] Add batch_flatten operator
• #18690 - Add NN operator attributes include to TensorRT codegen

Runtime

• #19476 - [REFACTOR]Phase out include/tvm/runtime/object.h
• #19465 - [REFACTOR][CODEGEN] Backend specific target and runtime to enable cross-compile fallback
• #19471 - [REFACTOR]Phase out IntTuple alias; use ffi::Shape directly
• #19472 - [REFACTOR]Phase out include/tvm/runtime/builtin_fp16.h
• #19469 - [REFACTOR]Phase out include/tvm/runtime/threading_backend.h
• #19455 - [REFACTOR]Phase out profiling.h heavy types, rename to timer.h
• #19457 - [REFACTOR]Macro cleanup — TVM_DLL alignment, [[maybe_unused]], logging.h legacy macros
• #18837 - [Builtin] Handle mismatched type on argument #0 when calling Builtin Runtime Operators
• #18813 - [REFACTOR]Phase out legacy contrib runtime backends
• #18784 - [REFACTOR]Transition metadata into ffi
• #18756 - [COMPACT] Fix 32bit compact in vm

TOPI

• #18880 - Reject non-float inputs for inverse unary math ops

TVMScript

• #18891 - Remove T.Bind backward-compat alias
• #18889 - Normalize T.Bind to T.bind for statement builder convention
• #18856 - Fix PEP 563 closure variable resolution

Vulkan

• #18914 - Avoid explicit layout decoration on non-interface allocations

web

• #18944 - Update includes after FFI JSON refactor
• #18893 - [Experimental] Add support for cross-origin storage caching
• #18680 - [Version] Fix WebLLM vision model issues
• #18687 - Handle LocalSession init in WASM RPC server

Misc

• #19483 - [REFACTOR][FFI] Cleanup ffi indirections in tvm headers + switch logging.h to ffi/error.h where only ICHECK/THROW are used
• #19484 - [FFI][ABI] Bump tvm-ffi to 0.1.11rc2
• #19477 - [REFACTOR] Delete src/support/libinfo.cc; replace with runtime FFI-registry env query
• #19479 - [REFACTOR][SCRIPT] TVMScript dialect-friendly refactor: per-dialect restructure + dialect registry
• #19475 - [REFACTOR][S-TIR] Move tvm/support/random_engine.h → tvm/s_tir/random_engine.h
• #19474 - [REFACTOR][IR] Move tvm/support/with.h → tvm/ir/with_context.h
• #19453 - [S-TIR][Dlight] Add layered fall back strategy to handle missing attr max_shared_memory_per_block
• #19463 - [REFACTOR][IR] Migrate include/tvm/node into include/tvm/ir
• #19462 - [REFACTOR][NODE] Use fn_repr inside kRepr lambdas, not ffi::ReprPrint
• #19460 - [REFACTOR][S-TIR] Minimize src/support/ by relocating s_tir-private headers
• #19459 - [REFACTOR] Phase out src/support/ffi_testing.cc
• #19461 - [REFACTOR][NODE] Migrate ReprPrinter to tvm-ffi ffi_repr mechanism
• #19456 - [REFACTOR] Move source_utils.h into runtime/opencl
• #19458 - [REFACTOR] Phase out unreachable contrib/rust_extension.cc
• #19454 - [REFACTOR][CODEGEN] Phase out tvm_global_barrier_state and tvm_prepare_global_barrier
• #19449 - [REFACTOR] Use FFI types in runtime inline module-create wrapper signatures
• #18406 - [TIR] Update symbolic index term order in loop fusion
• #19447 - [REFACTOR] Isolate backend module creation via ffi.Module.create. registry
• #19444 - [CMAKE][REFACTOR] Split libtvm.so into libtvm_runtime.so and libtvm_compiler.so
• #19440 - [REFACTOR] Remove runtime/object.py shim and route Object via tvm_ffi
• #19442 - [REFACTOR] Remove tvm.runtime.packed_func and container shims; route via tvm_ffi
• #19441 - [REFACTOR] Phase out include/tvm/runtime/module.h
• #19393 - fix: use is None instead of == None in test files (PEP 8 E711)
• #19406 - [S-TIR] Fix cache_read/cache_write region when inner block has T.whe…
• #19403 - [S-TIR] Fix Segfault when applying Parallel during TIR schedule rewriting
• #18927 - feat(meta_schedule): expand CUDA unroll steps for SM70 optimization
• #19347 - fix: TFLite model retrieval with error handling
• #19343 - test(relax): cover TFLite LOG and GREATER_EQUAL in test_frontend_tflite
• #18938 - [FFI] Bump tvm-ffi to 63224e3 and fix regressions
• #18912 - [TIR] Handle Bind in LowerDeviceKernelLaunch
• #18926 - Revert "fix: add safety warning to pickle_memoize cache loading"
• #18925 - fix: add safety warning to pickle_memoize cache loading
• #18913 - [Refactor] Bring up tirx namespace
• #18240 - [Optimization][Operator] Implement and enable Conv2d-Reshape-Add-ReLU fusion
• #18892 - [Build] Fix version regex to anchor at line start in pyproject.toml
• #18879 - [TIR] Reject non-floating inputs for trig unary ops
• #18886 - [TIR][REFACTOR] Revamp Common Subexpression Elimination
• #18883 - [TARGET] Fix round-trip reconstruction of targets with canonicalizer-generated feature.* attrs
• #18876 - [REFACTOR][TIR] Remove body from AllocBuffer and DeclBuffer
• #18871 - Batched GPU dispatch and object caching for WebGPU runtime
• #18875 - [chore] Update docker/README.md documentation and fix links
• #18873 - [TIR] Add VisitBufferDef/VisitBufferUse to base StmtVisitor/StmtMutator
• #18865 - [REFACTOR][TIR] Introduce AllocBuffer and phase out Allocate+DeclBuffer
• #18862 - [REFACTOR][TIR] Cleanup AttrStmt attributes
• #18857 - [TIR][Refactor] Enhance error reporting with structured AssertStmt and TVMFFIABIBuilder
• #18861 - fix: Complete CHECK update across contrib runtime
• #18859 - fix: Use T.decl_buffer instead of T.Buffer for aliased buffers in LongRoPE
• #18858 - fix: Complete ICHECK update across codebase
• #18845 - [REFACTOR][CONTRIB] Remove MSC contrib module
• #18847 - [PYTHON] Fix PEP 563 compat and remove args_converter
• #18843 - [TIR][FEAT] Require DeclBuffer before use in verify_well_formed
• #18852 - [REFACTOR] Remove unused mscclpp contrib module
• #18849 - [CMAKE] Remove unused Libbacktrace.cmake
• #18844 - [REFACTOR] Further cleanup node redirections
• #18830 - [LINT][PYTHON] Modernize annotations with ruff UP rules
• #18832 - [IR][TIR] Remove body from AssertStmt
• #18829 - [REFACTOR][NODE] Remove node redirect headers
• #18825 - [REFACTOR] Update CHECK and ICHECK_GE to TVM_FFI_ICHECK and TVM_FFI_ICHECK_GE in thrust.cu
• #18828 - [REFACTOR] Phase out root Makefile
• #18821 - fix: replace 6 bare except clauses with except Exception
• #18822 - [TARGET] Specify correct mcpu for Metal target tags
• #18816 - [REFACTOR][S-TIR] Lift STIR-only attributes out of tir::attr namespace
• #18810 - [REFACTOR][LINT] Modernize ruff config
• #18801 - Bump tvm-ffi to v0.1.9rc
• #18807 - [LINT] Modernize lint to use pre-commit hooks
• #18803 - [REFACTOR] Migrate CHECK macros to tvm-ffi ones
• #18768 - support integer types in fast_tanh and fast_exp
• #18802 - [FFI] Bring up latest tvm-ffi
• #18793 - [REFACTOR][TARGET] Further cleanup target python api
• #18785 - [REFACTOR][TARGET] Phase out legacy target string in favor of json
• #18786 - [CONTRIB] Cache the shape and dtype array in json access
• #18781 - [chore] cleanup unsed legacy backtrac code in logging
• #18779 - [REFACTOR] Phase out dmlc dep
• #18776 - [REFACTOR][S-TIR] More migrations to s-tir
• #18771 - [REFACTOR][S-TIR] Migrate more transform to s_tir
• #18763 - [REFACTOR][TIR] Phaseout BufferRealize
• #18759 - [REFACTOR] Remove picojson dependency, replace with tvm::ffi::json API
• #18760 - [chore] Cleanup stale dependencies
• #18761 - [REFATOR][TIR] Phase out AllocConst
• #18758 - [Cleanup] Remove redundant python/pyproject.toml and gen_requirements
• #18697 - Fix Customize Optimization tutorial import error [Bug] Customize Optimization Tutorial Error #18584
• #18754 - [REFACTOR][S-TIR] Cleanup items on block scope
• #18743 - [REFACTOR][S-TIR] Move remaining data structures to s_tir
• #18739 - fix: correct typos 'recieve' and 'occurence'
• #18744 - fix: skip dsymutil for static tvm_runtime on Apple platforms
• #18740 - fix: correct typo 'occuring' to 'occurring'
• #18738 - [chore][TIR] reorganize src/tir/transforms to src/tir/transform
• #18734 - [REFACTOR][S-TIR] Lift dlight into s_tir namespace
• #18735 - [REFACTOR][S-TIR] Migrate meta_schedule into s_tir namespace
• #18705 - Add Windows-specific build notes to installation guide
• #18727 - fix: correct typos in Python docstrings
• #18728 - [REFACTOR][S-TIR] Migrate tir/schedule to s_tir
• #18722 - [REFACTOR][S-TIR] Lift transform passes to s_tir namespace
• #18724 - Remove cron schedule from nightly Docker update workflow
• #18716 - [REFACTOR] Migrate old tir.ir_builder to tvmscript or builder
• #18715 - [SPIRV] Fix forloop codegen in vulkan
• #18712 - [REFACTOR][S-TIR] Initialize the s_tir module
• #18636 - [TIR][Schedule]Generalize fuseReductionEpilogue to support arbitrary epilogue expressions
• #18699 - [TIR] Further robustify floordiv/mod intrin lowering to prevent overflow
• #18689 - [REFACTOR][TIR] Rename tir.Block to SBlock
• #18671 - [TIR] Fix InjectPTXLDG32 segfaults and skip non-CUDA targets