Commits

Branch selector

 main

User selector

All users

All time

Commit History

Commits on Apr 23, 2026

• [Relax][FRONTEND][ONNX] Support Softmax, LogSoftmax and Hardmax when opset version ≤12 (#19428)

  Show description for 7eea6df
  cchung100m

  authored
  7eea6df

  Copy full SHA for 7eea6df

Commits on Apr 22, 2026

• [FIX] Skip metal target tag registration for unsupported LLVM CPUs (#19427)

  tqchen

  authored
  a8f1ace

  Copy full SHA for a8f1ace

Commits on Apr 21, 2026

• [BugFix][LLVM] Fix `insertDeclare` API mismatch for ROCm-bundled LLVM 20 (#19390)

  Show description for d7282a3
  swjng

  authored
  d7282a3

  Copy full SHA for d7282a3

• [Backend][Relax] Add NPU BYOC backend example (#19425)

  Show description for 21993a5
  Aristide021

  authored
  21993a5

  Copy full SHA for 21993a5

Commits on Apr 20, 2026

• [Relax] Fix deprecation warning (#19424)

  Show description for 527fbf5
  cchung100m

  authored
  527fbf5

  Copy full SHA for 527fbf5

Commits on Apr 18, 2026

• [Relax][TVMScript] Print ExternFunc struct_info when non-default (#19416)

  Show description for ca68bef
  cchung100m

  authored
  ca68bef

  Copy full SHA for ca68bef

• [Relax][Frontend][TFLite] Fix bool `REDUCE_ANY`/`REDUCE_ALL` compile failure (#19415)

  Show description for 545c332
  swjng

  authored
  545c332

  Copy full SHA for 545c332

Commits on Apr 17, 2026

• [Docs] Fix stale tvm.tirx exclude list and add missing legalize_ops.unary entry (#19414)

  Show description for ff127c4
  tlopex

  authored
  ff127c4

  Copy full SHA for ff127c4

• [Relax][Frontend][TFLite] Add `REDUCE_ANY` and `REDUCE_ALL` (#19413)

  Show description for f0af322
  swjng

  authored
  f0af322

  Copy full SHA for f0af322

• [relax] fix 'occured' -> 'occurred' in transform.h doc comment (#19411)

  Show description for 379bd1f
  SAY-5

  authored
  379bd1f

  Copy full SHA for 379bd1f

• fix: use `is None` instead of `== None` in test files (PEP 8 E711) (#19393)

  Show description for 0fdb2cd
  

  harshadkhetpal

  and

  MasterJH5574

  authored
  0fdb2cd

  Copy full SHA for 0fdb2cd

Commits on Apr 16, 2026

• [Relax][Frontend][TFLite] Fix and test `MATRIX_DIAG`, `MATRIX_SET_DIAG`, `SPARSE_TO_DENSE` (#19408)

  Show description for 75a6b30
  Aharrypotter

  authored
  75a6b30

  Copy full SHA for 75a6b30

• [Fix][Runtime][RPC] Fix remote tensor handle cleanup for RPC return values (#19410)

  Show description for 5c17111
  tlopex

  authored
  5c17111

  Copy full SHA for 5c17111

• [Docs] Fix outdated source install and API reference docs (#19409)

  Show description for e0e9315
  tlopex

  authored
  e0e9315

  Copy full SHA for e0e9315

• [S-TIR] Fix cache_read/cache_write region when inner block has T.whe… (#19406)

  Show description for 0a79095
  elvin-n

  authored
  0a79095

  Copy full SHA for 0a79095

• [S-TIR] Fix Segfault when applying Parallel during TIR schedule rewriting (#19403)

  Show description for 9d13fc0
  cchung100m

  authored
  9d13fc0

  Copy full SHA for 9d13fc0

Commits on Apr 15, 2026

• [Relax][Frontend][KVCache] Restructure kv_cache kernels (#19405)

  Show description for b343943
  MasterJH5574

  authored
  b343943

  Copy full SHA for b343943

• Fix #18714: [Docs] python -c "import tvm; print(tvm.file)" fail (#19407)

  Show description for 1b94055
  

  JiwaniZakir

  and

  claude

  authored
  1b94055

  Copy full SHA for 1b94055

• [relax][tflite] Add PRELU/LRN/SQUARED_DIFFERENCE tests (partial #18971) (#19404)

  Show description for 14751b3
  0xjah

  authored
  14751b3

  Copy full SHA for 14751b3

Commits on Apr 14, 2026

• [Frontend][TFLite] Add test coverage for SHAPE and RANGE operators (#19401)

  Show description for 3d1e402
  

  rknastenka

  and

  gemini-code-assist[bot]

  authored
  3d1e402

  Copy full SHA for 3d1e402

• [Test][TFLite] Add unit tests for `PRELU` (#19402)

  Show description for 5efa4b7
  fnhirwa

  authored
  5efa4b7

  Copy full SHA for 5efa4b7

• [Relax][Frontend][KVCache] Add masked sequence prefill helper for encoder valid lengths (#19392)

  Show description for a6e2ea8
  xthomaswang

  authored
  a6e2ea8

  Copy full SHA for a6e2ea8

Commits on Apr 13, 2026

• [Docs] Add code generation architecture documentation (#19396)

  Show description for 87fdeaf
  tlopex

  authored
  87fdeaf

  Copy full SHA for 87fdeaf

• [Docs] Add TVMScript architecture documentation (#19398)

  Show description for f9787d1
  tlopex

  authored
  f9787d1

  Copy full SHA for f9787d1

• [Frontend][TFLite] Add TILE operator tests and edge cases (#19400)

  Show description for 60bb59d
  rknastenka

  authored
  60bb59d

  Copy full SHA for 60bb59d

• [Docs] Add PyModule tutorial to How-To toctree (#19397)

  Show description for 13867ea
  tlopex

  authored
  13867ea

  Copy full SHA for 13867ea

• [Docs] Clean up architecture docs: remove duplicates, fix stale content (#19399)

  Show description for e0babea
  tlopex

  authored
  e0babea

  Copy full SHA for e0babea

• feat(meta_schedule): expand CUDA unroll steps for SM70 optimization (#18927)

  Show description for b465646
  jianhua1724

  authored
  b465646

  Copy full SHA for b465646

Commits on Apr 12, 2026

• [Docs] Add Relax VM architecture documentation (#19389)

  Show description for 1d02ac3
  tlopex

  authored
  1d02ac3

  Copy full SHA for 1d02ac3

• [Docs] Add operator fusion architecture documentation (#19394)

  Show description for 481aa6e
  tlopex

  authored
  481aa6e

  Copy full SHA for 481aa6e

• [Docs] Add BYOC external library dispatch architecture documentation (#19395)

  Show description for e22be4e
  tlopex

  authored
  e22be4e

  Copy full SHA for e22be4e

• [relax][frontend][tflite] Add tests for fully_connected/depthwise_conv2d/transpose_conv/l2_pool2d (#19372)

  Show description for e5d4c55
  0xjah

  authored
  e5d4c55

  Copy full SHA for e5d4c55

• [Relax][ONNX] Add frontend support for QuantizeLinear, DequantizeLinear, and DynamicQuantizeLinear (#19391)

  Show description for 645fcf9
  hunghsiangwang

  authored
  645fcf9

  Copy full SHA for 645fcf9

Commits on Apr 11, 2026

• [BugFix][Relax] Select target-specific pipeline in tvm.compile when GPU target is provided (#19384)

  Show description for b9ced1a
  swjng

  authored
  b9ced1a

  Copy full SHA for b9ced1a

• [Docs] Add docstrings for nn.Module classes and core APIs in relax.frontend.nn (#19387)

  Show description for 4d812a1
  tlopex

  authored
  4d812a1

  Copy full SHA for 4d812a1