Issues

• [Bug] s_tir.LowerInitBlock skips T.init() when symbolic reduction extent is 0, returning uninitialized memory

  apache/tvm#20974 · Filter by author xgbtxy opened on Nov 11, 2026

• [Bug] [TIRx] CompactBufferAllocation leaves a stale double_buffer_scope reference and causes InjectDoubleBuffer to fail

  needs-triagetype: bug
  apache/tvm#20972 · Filter by author pinpinpoo opened on Nov 11, 2026

• [Feature][Relax][ONNX] Support CenterCropPad operator in Relax ONNX frontend

  apache/tvm#20971 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support MelWeightMatrix operator in Relax ONNX frontend

  apache/tvm#20970 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support LpNormalization operator in Relax ONNX frontend

  apache/tvm#20969 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support DeformConv (Deformable Convolution) in Relax ONNX frontend

  apache/tvm#20968 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support RotaryEmbedding (RoPE) operator in Relax ONNX frontend

  apache/tvm#20967 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support STFT (Short-Time Fourier Transform) operator in Relax ONNX frontend

  apache/tvm#20966 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support ConvInteger operator in Relax ONNX frontend

  apache/tvm#20965 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support QLinearAveragePool operator in Relax ONNX frontend

  apache/tvm#20962 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support QLinearAdd operator in Relax ONNX frontend

  apache/tvm#20961 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support DFT (Discrete Fourier Transform) in Relax ONNX frontend

  apache/tvm#20960 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support QLinearConv in Relax ONNX frontend

  apache/tvm#20959 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Feature][Relax][ONNX] Support QLinearMatMul in Relax ONNX frontend

  apache/tvm#20958 · Filter by author Chaitanya140904 opened on Nov 10, 2026

• [Performance] image.resize2d nearest (asymmetric, non-floor rounding) is ~2x slower on CUDA since v0.24.0 (int64 index math from #18838)

  apache/tvm#20956 · Filter by author junghyunpark2001 opened on Nov 10, 2026

• [Bug][Relax][ONNX] BitShift raises AttributeError: 'str' object has no attribute 'decode' when direction attribute is omitted

  needs-triagetype: bug
  apache/tvm#20950 · Filter by author Anirudh-65 opened on Nov 9, 2026

• [Bug][Relax][ONNX] SequenceLength, SequenceErase, and SequenceInsert fail with "TypeError: object of type 'Call' has no len()"

  needs-triagetype: bug
  apache/tvm#20949 · Filter by author Anirudh-65 opened on Nov 9, 2026

• [Bug][Relax][PyTorch] torch.prod with reduction dimension fails in from_exported_program: "AssertionError: Unsupported function types ['prod.dim_int']"

  needs-triagetype: bug
  apache/tvm#20948 · Filter by author Anirudh-65 opened on Nov 9, 2026

• [Bug][Relax][PyTorch] Boolean mask indexing fails in from_exported_program: "TypeError: Take op requires the input indices to have integer dtype. However, the given indices dtype is T.bool"

  needs-triagetype: bug
  apache/tvm#20947 · Filter by author Anirudh-65 opened on Nov 9, 2026

• [Bug][Relax] Arithmetic on an unbatched AdaptiveAvgPool2d output fails in from_exported_program: "Please convert 2.0 with const first"

  needs-triagetype: bug
  apache/tvm#20943 · Filter by author raindrops-0199 opened on Nov 9, 2026
  1

• [Bug][Relax] torch.max(x, dim) on a bool tensor fails at runtime: "Unsupported input dtype: bool"

  needs-triagetype: bug
  apache/tvm#20942 · Filter by author raindrops-0199 opened on Nov 9, 2026
  1

• [Bug][Relax] interpolate(mode="bilinear", align_corners=True) to an output size of 1 fails in tvm.compile: "Divide by zero"

  needs-triagetype: bug
  apache/tvm#20941 · Filter by author raindrops-0199 opened on Nov 9, 2026

• [Bug][Relax] torch.where with branches of different dtypes fails in from_exported_program

  needs-triagetype: bug
  apache/tvm#20940 · Filter by author raindrops-0199 opened on Nov 9, 2026
  1

• [Bug][Relax] x * t.item() fails in from_exported_program when x is float and t is an integer tensor

  needs-triagetype: bug
  apache/tvm#20939 · Filter by author raindrops-0199 opened on Nov 9, 2026

• [Bug] [Relax][ONNX] Float8 Cast drops saturate and merges distinct programs

  needs-triagetype: bug
  apache/tvm#20936 · Filter by author sunni13830 opened on Nov 7, 2026