Google Git
Sign in
apache/tvm/HEAD
  1. 726edbf [Docs] Complete API reference for tvm.relax backend and testing modules (#19378) by Shushi Hong · 12 hours ago main nightly
  2. 5d6c972 [BugFix][Relax] Fix ONNX `get_converter` index underflow for old opsets (#19376) by Soowon Jeong · 18 hours ago
  3. 4f5a17a [BugFix][Relax] Fix ONNX Clip converter for opset 11-12 (#19375) by Soowon Jeong · 22 hours ago
  4. fb6453a [Relax][TFLite] Fix `MIRROR_PAD`/`ONE_HOT` converters and add tests for `PAD`, `PADV2`, `MIRROR_PAD`, `TOPK_V2`, `ONE_HOT` (#19373) by Soowon Jeong · 22 hours ago
  5. 7dcdb56 [Docs] Add API reference for tvm.s_tir submodules: dlight, meta_schedule, backend (#19369) by Shushi Hong · 24 hours ago
  6. 643cf60 [Relax][TFLite] Add test coverage for Reduction operations (#18971) (#19370) by Bana · 2 days ago
  7. 2e6ee08 [BugFix] Align `tir.round` to ties-to-even across all backends (#19368) by Soowon Jeong · 2 days ago
  8. 2345e6e [Docs] Add API reference documentation for tvm.script module (#19366) by Shushi Hong · 2 days ago
  9. 865ace9 [Docs] Add DLight and MetaSchedule deep-dive instructions (#19356) by Shushi Hong · 2 days ago
  10. 9acbf4a [BugFix][Relax] Add structural_equal verification to subroutine cache lookup (#18962) by 3em0 · 3 days ago
  11. acfc483 [BugFix][ONNX] Fix Round op to use ties-to-even (#19367) by Soowon Jeong · 3 days ago
  12. f0863fd [Relax][ONNX] Support ConcatFromSequenc/SequenceInsert with new_axis=1 (#19361) by Bana · 4 days ago
  13. 072c849 [Test][TFLite] Add unit tests for RESIZE_BILINEAR and RESIZE_NEAREST_NEIGHBOR ops (#19365) by Kryptonite · 4 days ago
  14. e0df802 [Docs] TFLite tests requiring Python 3.10 and specific package versions to avoid core dumps (#19364) by Bana · 4 days ago
  15. d58e9f7 [WebGPU] Add gating logic for subgroup shuffle primitives (#18823) by Sai Gopal Reddy Kovvuri · 4 days ago
  16. d1f5583 [Docs] Add tutorial for importing models from PyTorch, ONNX, and TFLite (#19354) by Shushi Hong · 4 days ago
  17. 93e28d1 [BugFix][TVMScript] Fix invalid f-string format spec causing TypeError on Python 3.14 (#19362) by Soowon Jeong · 4 days ago
  18. 9ec6a52 [Docs] Add Dataflow Pattern Language (DPL) documentation for Relax (#19358) by Shushi Hong · 4 days ago
  19. 628b394 [Docs] Add Disco distributed runtime architecture overview (#19357) by Shushi Hong · 4 days ago
  20. 744ef56 [Docs] Fix outdated paths, links, and add missing API references across documentation (#19351) by Shushi Hong · 4 days ago
  21. 491480d [Docs] Add tvm.s_tir.analysis API reference page (#19353) by Shushi Hong · 4 days ago
  22. 5673754 Bump tvm-ffi to 1fed0a (#18967) by Yong Wu · 4 days ago
  23. 36a82f5 [BugFix][TVMScript] Add `doc.keyword` handling for `ExprEvaluator._visit` (#19352) by liggest · 5 days ago
  24. b8ecf2f [Docs] Add Relax VM architecture overview in documentation (#19350) by Shushi Hong · 6 days ago
  25. befee58 [Docs] Fix outdated code examples, typos, and missing API reference in documentation(2) (#19344) by Shushi Hong · 6 days ago
  26. 28aac47 [Relax][TFLite] Add NON_MAX_SUPPRESSION_V5 support (#19349) by Bana · 6 days ago
  27. 78b7286 fix: TFLite model retrieval with error handling (#19347) by Bana · 6 days ago
  28. 87f2915 test(relax): cover TFLite LOG and GREATER_EQUAL in test_frontend_tflite (#19343) by Dayuxiaoshui · 6 days ago
  29. e3dda23 [TFLite][Frontend] Add expected IRModule checks for conv2d, pool2d, and batch_matmul tests (#18970) by Shushi Hong · 7 days ago
  30. 6601962 [Docs] Fix outdated code examples, types, and missing references across documentation (#18965) by Shushi Hong · 7 days ago
  31. 427b66d [Frontend][ONNX] Fix SplitToSequence keepdims=0 and uneven last chunk (#19341) by Kryptonite · 7 days ago
  32. 10ba3c2 [Frontend][ONNX] Support select_last_index for ArgMax and ArgMin (#18969) by Kryptonite · 8 days ago
  33. ccb84cd [Web] Fix rollup errors and bump tvmjs version (#18958) by Akaash Parthasarathy · 8 days ago
  34. 81889de [Fix] Replace str(target.kind) with target.kind.name for Target objects (#18959) by Akaash Parthasarathy · 8 days ago
  35. ad65d9f [WebGPU] Reserve additional keywords to avoid WGSL identifier collisions (#18960) by Akaash Parthasarathy · 8 days ago
  36. eb53118 [DOC] Fix various issues (#18966) by Ruslan Baratov · 8 days ago
  37. a7bfc85 [FIX] Inline ceil_log2 in gpu_2d_continuous_cumsum to fix MakePackedAPI error (#18957) by Gabe Guralnick · 9 days ago
  38. ec0daad [Relax][ONNX] Support Resize dynamic ROI via TOPI (#18963) by Dayuxiaoshui · 9 days ago
  39. fd9d9db [Relax][ONNX] Fix shape/dynamic restrictions for `Squeeze`/`Unsqueeze` and `Slice` (#18955) by HoYi · 10 days ago
  40. d293b7a [Docs] Align documentation with tirx/s_tir namespace split (#18953) by Shushi Hong · 11 days ago
  41. c79caf0 [Relax][ONNX] Complete ShapeExpr reshape handling in ONNX frontend (#18956) by YinHanke · 11 days ago
  42. cb5e290 [Relax][ONNX] Add Optional and MatMulInteger16 frontend support (#18950) by HoYi · 11 days ago
  43. e229bda [Docs] Add tutorial for mixing Python/PyTorch with TVM using BasePyModule (#18947) by Shushi Hong · 12 days ago
  44. 8597d21 [Frontend][ONNX] Add MatMulInteger support to Relax ONNX frontend (#18951) by Kryptonite · 12 days ago
  45. 4df6b17 [Relax][ONNX] Add roi_pool op and MaxRoiPool frontend support (#18952) by YinHanke · 12 days ago
  46. 4de1f11 [Relax] Add conv3d_transpose and ONNX ConvTranspose 3D support (#18948) by Dayuxiaoshui · 12 days ago
  47. 52b5d55 [Frontend][ONNX] Add If operator support to Relax ONNX frontend (#18946) by Kryptonite · 13 days ago
  48. 38eb79c [Relax][Vision] Add get_valid_counts and classic NMS (#18943) by HoYi · 13 days ago
  49. 9dcc731 [Web] Update includes after FFI JSON refactor (#18944) by Akaash Parthasarathy · 13 days ago
  50. 44dbd13 [FFI] Bump tvm-ffi to 63224e3 and fix regressions (#18938) by Tianqi Chen · 13 days ago
  51. 3eb86f7 [Relax][TOPI] Add relax.vision.multibox_transform_loc for SSD/TFLite box decode (#18942) by Dayuxiaoshui · 14 days ago
  52. f66c048 [DOC] Fix inconsistent code comments (#18939) by cj · 14 days ago
  53. 9c8e5a6 [TIR] Handle Bind in LowerDeviceKernelLaunch (#18912) by Ruihang Lai · 2 weeks ago
  54. 53b6576 [Fix] Fix tvm.tir references in Tflite frontend (#18940) by Shushi Hong · 2 weeks ago
  55. 90e6e8b [Docs] Fix duplicate license headers and incorrect module paths after tirx rename (#18941) by Shushi Hong · 2 weeks ago
  56. 2f2469e [Relax] Add affine_grid operator with PyTorch and ONNX frontend support (#18933) by HoYi · 2 weeks ago
  57. bf6ed31 [Relax][PyTorch] Add 3D interpolate support using resize3d (#18937) by Nirdesh Devadiya · 2 weeks ago
  58. 1e08eb2 [Relax][ONNX][Torch] Add roi_align support and frontend integration (#18936) by YinHanke · 2 weeks ago
  59. 4d36a45 [Relax][ONNX] Add image.resize3d op and wire 5D Resize (#18931) by Dayuxiaoshui · 2 weeks ago
  60. 59c14f6 [Relax][ONNX] Add GridSample ONNX frontend integration (#18932) by Kryptonite · 2 weeks ago
  61. e53cfe1 [Frontend][TFLite] Fix undefined symbols and Relay API remnants in TFLite frontend (#18929) by Shushi Hong · 2 weeks ago
  62. ff883db Revert "fix: add safety warning to pickle_memoize cache loading" (#18926) by Tianqi Chen · 2 weeks ago
  63. 7b3fa38 fix: add safety warning to pickle_memoize cache loading (#18925) by scruge1 · 2 weeks ago
  64. 5e05f70 [Relax][TFLite] Introduce TensorFlow Lite frontend (#18868) by Siva · 3 weeks ago
  65. d33630c [Vulkan] Avoid explicit layout decoration on non-interface allocations (#18914) by Yu Chengye · 3 weeks ago
  66. 00eb226 [Refactor] Update type references from tir to tirx in PyTorch ExportedProgram frontend (#18920) by Masahiro Hiramori · 3 weeks ago
  67. c581abe [Relax][PyTorch] Fix _slice and _expand for dynamic shapes in PyTorch ExportedProgram frontend (#18918) by Masahiro Hiramori · 3 weeks ago
  68. 69336ac [TOPI] Fix strided_slice_with_axes to handle negative axis values (#18917) by Masahiro Hiramori · 3 weeks ago
  69. 141c22f [Refactor] Bring up tirx namespace (#18913) by Tianqi Chen · 3 weeks ago
  70. c9fb8cd [Docs] Clean up stale references from recent refactors (#18908) by Shushi Hong · 3 weeks ago
  71. 42a5349 [Target][LLVM] Fix -mcpu validation compatibility across LLVM versions (#18909) by Shushi Hong · 3 weeks ago
  72. be2ea89 [Docs] Update outdated references from recent refactors (#18906) by Shushi Hong · 4 weeks ago
  73. 474cde4 [Optimization][Operator] Implement and enable Conv2d-Reshape-Add-ReLU fusion (#18240) by kimm240 · 4 weeks ago
  74. 353cc70 [Web][Experimental] Add support for cross-origin storage caching (#18893) by Akaash Parthasarathy · 4 weeks ago
  75. 64a24c4 [RELAX][LAYOUT] Support multiple axis paching (#18869) by Siva · 4 weeks ago
  76. 3788e99 [Relax][PyTorch] Add torch.cond support to ExportedProgram frontend (#18904) by Masahiro Hiramori · 4 weeks ago
  77. 122592b [Relax] Add input type validation for make_shape and corresponding tests (#18870) by anusha975 · 4 weeks ago
  78. 1499bda [Relax][PyTorch] Fix crash on dynamic shapes with identity slice in ExportedProgram importer (#18903) by Masahiro Hiramori · 4 weeks ago
  79. 7a41af0 [FIX][LLVM] Use isCPUStringValid for mcpu validation instead of enumerating processor descriptions (#18884) by Gabe Guralnick · 4 weeks ago
  80. 728eb15 [Build] Fix version regex to anchor at line start in pyproject.toml (#18892) by Ruihang Lai · 5 weeks ago
  81. 87bd8af [TVMScript] Remove T.Bind backward-compat alias (#18891) by Tianqi Chen · 5 weeks ago
  82. a89b9f2 [TOPI] Reject non-float inputs for inverse unary math ops (#18880) by YinHanke · 5 weeks ago
  83. 5d3b525 [TIR] Reject non-floating inputs for trig unary ops (#18879) by YinHanke · 5 weeks ago
  84. f83cebb [TVMScript] Normalize T.Bind to T.bind for statement builder convention (#18889) by Tianqi Chen · 5 weeks ago
  85. 72de122 [TIR][REFACTOR] Revamp Common Subexpression Elimination (#18886) by Tianqi Chen · 5 weeks ago
  86. 717c822 [FIX] Fix cumsum kernel sblock_alloc_buffer for non-sblock buffer (#18887) by Tianqi Chen · 5 weeks ago
  87. 9dfa116 [Metal] Batched command dispatch and staging buffer pool (#18877) by Miti · 5 weeks ago
  88. c0a305d [TARGET] Fix round-trip reconstruction of targets with canonicalizer-generated `feature.*` attrs (#18883) by Masahiro Hiramori · 5 weeks ago
  89. 17d1a28 [FIX][Adreno] Replace AllocBuffer with Bind in texture alloc injection (#18881) by Tianqi Chen · 5 weeks ago
  90. 689d2b5 [REFACTOR][TIR] Remove body from AllocBuffer and DeclBuffer (#18876) by Tianqi Chen · 5 weeks ago
  91. 14e41c6 [Relax][ONNX] Support dynamic repeats for Tile (#18878) by YinHanke · 5 weeks ago
  92. 419a8c86 Batched GPU dispatch and object caching for WebGPU runtime (#18871) by Miti · 5 weeks ago
  93. dd7e260 [chore] Update docker/README.md documentation and fix links (#18875) by Zhengke Zhou · 5 weeks ago
  94. 079e4af [REFACTOR][TIR] Rename LetStmt to Bind and flatten to sequential semantics (#18874) by Tianqi Chen · 5 weeks ago
  95. 969fad3 [TIR] Add VisitBufferDef/VisitBufferUse to base StmtVisitor/StmtMutator (#18873) by Tianqi Chen · 5 weeks ago
  96. 0fba160 [REFACTOR][TIR] Introduce AllocBuffer and phase out Allocate+DeclBuffer (#18865) by Tianqi Chen · 5 weeks ago
  97. 21e5225 [ADRENO] Revive and consolicate Adreno features (#18867) by Siva · 5 weeks ago
  98. 75f8589 [CI] Update images to `20260301-134651-63f099ad` (#18827) by Masahiro Hiramori · 5 weeks ago
  99. 521440e [REFACTOR][TIR] Cleanup AttrStmt attributes (#18862) by Tianqi Chen · 6 weeks ago
  100. c481950 [Relax][Refactor] Phase out FewShotTuning (#18864) by Tianqi Chen · 6 weeks ago