commits  

Messages by Thread

• • Re: [PR] [Relax][PyTorch] Support gru op for ExportedProgram importer [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Support gru op for ExportedProgram importer [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Support gru op for ExportedProgram importer [tvm] via GitHub
• (tvm-ffi) branch main updated: feat: Metadata and JSON-based type schemas for fields, methods and global funcs (#36) tqchen
• (tvm-ffi) branch main updated: [CYTHON] Fix memory leak when tensor alloc returns to python (#87) tqchen
• (tvm) branch nightly updated (d9ced6e56a -> 6c37194c2f) github-bot
• [PR] [Relax][PyTorch] Fix the segfault in from_exported_program when model returns (Tensor, None) tuple [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Fix the segfault in from_exported_program when model returns (Tensor, None) tuple [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Fix the segfault in from_exported_program when model returns (Tensor, None) tuple [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Fix the segfault in from_exported_program when model returns (Tensor, None) tuple [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Fix the segfault in from_exported_program when model returns (Tensor, None) tuple [tvm] via GitHub
• [PR] [Relax][Frontend][ONNX] Fix `FastGelu` when bias does not set [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][ONNX] Fix `FastGelu` when bias does not set [tvm] via GitHub
• [I] [bug][relax.frontend.torch] from_exported_program KeyError for non-persistent buffer bert.embeddings.position_ids (HuggingFace BERT via torch.export) [tvm] via GitHub
  • Re: [I] [bug][relax.frontend.torch] from_exported_program KeyError for non-persistent buffer bert.embeddings.position_ids (HuggingFace BERT via torch.export) [tvm] via GitHub
• [I] [Bug] [feature][relax.frontend.torch] from_exported_program rejects fused GRU (aten::gru.input) with “Unsupported function types ['gru.input']” [tvm] via GitHub
  • Re: [I] [Bug] [feature][relax.frontend.torch] from_exported_program rejects fused GRU (aten::gru.input) with “Unsupported function types ['gru.input']” [tvm] via GitHub
  • Re: [I] [Bug] [feature][relax.frontend.torch] from_exported_program rejects fused GRU (aten::gru.input) with “Unsupported function types ['gru.input']” [tvm] via GitHub
• [I] [Bug][relax.frontend.torch] Segfault in relax::Tuple when importing/building a torch.export program that uses batch-wise advanced indexing write (aten.index_put_) [tvm] via GitHub
• (tvm) branch nightly updated (6c37194c2f -> 3015acd7e6) github-bot
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@3015acd7e678cd97e4334835e130c09b315ab1a7) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@3015acd7e678cd97e4334835e130c09b315ab1a7) tqchen
• (tvm) branch main updated: [FFI][ABI] Bump tvm-ffi to latest (#18354) tqchen
• Re: [I] [Bug] [FRONTEND][ONNX] fails to convert a valid onnx model since the Less operator cannot broadcast for valid inputs [tvm] via GitHub
  • Re: [I] [Bug] [FRONTEND][ONNX] fails to convert a valid onnx model since the Less operator cannot broadcast for valid inputs [tvm] via GitHub
• Re: [PR] added support for logsigmoid op [tvm] via GitHub
• (tvm) branch main updated: [Relax][ONNX] Support AllClassNMS Operator for ONNX Frontend (#18321) ruihangl
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@c00c66259a8dd4cf197601c978c566ce2db9bc17) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@c00c66259a8dd4cf197601c978c566ce2db9bc17) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@c00c66259a8dd4cf197601c978c566ce2db9bc17) tqchen
• (tvm-ffi) branch main updated: chore: fix lint issue (#84) yaxingcai
• (tvm-ffi) branch main updated: Bring a new beta version (#83) ruihangl
• (tvm-ffi) branch main updated: [TENSOR] Allow strides to be null for zerodim case (#82) tqchen
• [PR] [CUDA] Update FlashInfer JIT integration [tvm] via GitHub
  • Re: [PR] [CUDA] Update FlashInfer JIT integration [tvm] via GitHub
  • Re: [PR] [CUDA] Update FlashInfer JIT integration [tvm] via GitHub
  • Re: [PR] [CUDA] Update FlashInfer JIT integration [tvm] via GitHub
• (tvm-ffi) branch main updated: [ABI] Stablize doc and metadata for functions (#79) ruihangl
• (tvm-ffi) branch main updated: feat: add trainium device (#76) tqchen
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@6c37194c2f38b505638ea06404b1e523da068456) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@6c37194c2f38b505638ea06404b1e523da068456) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@6c37194c2f38b505638ea06404b1e523da068456) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@6c37194c2f38b505638ea06404b1e523da068456) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@6c37194c2f38b505638ea06404b1e523da068456) tqchen
• (tvm-ffi) branch main updated: [ABI] Further clarify Function ABI in cpp (#71) yongwww
• (tvm-ffi) branch main updated: [ERROR] Place ErrorKind second in error handling (#70) ruihangl
• (tvm) branch nightly updated (f68651f035 -> d9ced6e56a) github-bot
• (tvm) branch main updated: [Relax][Frontend][Torch] Fix parsing error when input dimension of unbind is 1 (#18351) tlopex
• [PR] [Fix] Update ShapeView use in nccl.cc [tvm] via GitHub
  • Re: [PR] [Fix] Update ShapeView use in nccl.cc [tvm] via GitHub
  • Re: [PR] [Fix] Update ShapeView use in nccl.cc [tvm] via GitHub
  • Re: [PR] [Fix] Update ShapeView use in nccl.cc [tvm] via GitHub
• [PR] [Relax][Frontend][Torch] Fix parsing error when input dimension of unbind is 1 [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][Torch] Fix parsing error when input dimension of unbind is 1 [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][Torch] Fix parsing error when input dimension of unbind is 1 [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][Torch] Fix parsing error when input dimension of unbind is 1 [tvm] via GitHub
  • Re: [PR] [Relax][Frontend][Torch] Fix parsing error when input dimension of unbind is 1 [tvm] via GitHub
• [I] [Bug] Did you forget to bind? for cuda backend. [tvm] via GitHub
  • Re: [I] [Bug] Did you forget to bind? for cuda backend. [tvm] via GitHub
  • Re: [I] [Bug] Did you forget to bind? for cuda backend. [tvm] via GitHub
  • Re: [I] [Bug] Did you forget to bind? for cuda backend. [tvm] via GitHub
  • Re: [I] [Bug] Did you forget to bind? for cuda backend. [tvm] via GitHub
• (tvm) branch nightly updated (fc20c0a619 -> f68651f035) github-bot
• (tvm) branch main updated: [FFI][ABI] Bump tvm-ffi to latest (#18349) ruihangl
• (tvm-ffi) branch main updated: [CYTHON] Fix stream passing bug (#68) tqchen
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@2c7ef9427f35f9c3f8ae02c3cd589acfbaa96a5c) tqchen
• [PR] [FFI][ABI] Bump tvm-ffi to latest [tvm] via GitHub
  • Re: [PR] [FFI][ABI] Bump tvm-ffi to latest [tvm] via GitHub
  • Re: [PR] [FFI][ABI] Bump tvm-ffi to latest [tvm] via GitHub
  • [PR] [FFI][ABI] Bump tvm-ffi to latest [tvm] via GitHub
  • Re: [PR] [FFI][ABI] Bump tvm-ffi to latest [tvm] via GitHub
  • Re: [PR] [FFI][ABI] Bump tvm-ffi to latest [tvm] via GitHub
• (tvm) branch main updated (fc20c0a619 -> 2c7ef9427f) tlopex
• (tvm-ffi) branch main updated: [ABI] Introduce ShapeView Minimize TensorObj exposure (#67) tqchen
• (tvm-ffi) branch main updated (cfff30b -> 6fb42a7) junrushao
• [PR] [Python] Add library lookup path for tvm installed as a pakcage [tvm] via GitHub
  • Re: [PR] [Python] Add library lookup path for tvm installed as a pakcage [tvm] via GitHub
  • Re: [PR] [Python] Add library lookup path for tvm installed as a pakcage [tvm] via GitHub
  • Re: [PR] [Python] Add library lookup path for tvm installed as a pakcage [tvm] via GitHub
• [PR] tests(cuda): deterministic VerifyGPUCode test for shared-memory overflow (64KB vs 48KB) [tvm] via GitHub
  • Re: [PR] tests(cuda): deterministic VerifyGPUCode test for shared-memory overflow (64KB vs 48KB) [tvm] via GitHub
  • Re: [PR] tests(cuda): deterministic VerifyGPUCode test for shared-memory overflow (64KB vs 48KB) [tvm] via GitHub
  • Re: [PR] tests(cuda): deterministic VerifyGPUCode test for shared-memory overflow (64KB vs 48KB) [tvm] via GitHub
  • Re: [PR] tests(cuda): deterministic VerifyGPUCode test for shared-memory overflow (64KB vs 48KB) [tvm] via GitHub
  • Re: [PR] tests(cuda): deterministic VerifyGPUCode test for shared-memory overflow (64KB vs 48KB) [tvm] via GitHub
  • Re: [PR] tests(cuda): deterministic VerifyGPUCode test for shared-memory overflow (64KB vs 48KB) [tvm] via GitHub
• (tvm) branch nightly updated (36e473f58b -> fc20c0a619) github-bot
• (tvm-ffi) branch main updated: [VERSION] Expose version (#62) tqchen
• (tvm-ffi) branch main updated: fix: Only attach docstring when it's defined (#60) tqchen
• (tvm-ffi) branch main updated: [CYTHON] Fix potential flaky issue in dispatcher (#63) tqchen
• (tvm-ffi) branch main updated: test: `__doc__` being None when it's not explicitly set (#61) tqchen
• (tvm) branch main updated: [FFI][ABI] Bump tvm-ffi version to reflect RC ABI Update (#18345) tqchen
• (tvm-ffi) branch main updated: [CYTHON] Add always inline hint to python helpers (#59) tqchen
• (tvm-ffi) branch tensor-abi deleted (was c708337) tqchen
• (tvm-ffi) branch tensor-abi created (now c708337) tqchen
  • (tvm-ffi) 01/01: ABI hack tqchen
• (tvm) branch docs-check created (now 195eb0c856) tqchen
  • (tvm) 01/01: DOCS check tqchen
• (tvm) branch nightly updated (a21e0df4b3 -> 36e473f58b) github-bot
• [PR] [Relax][PyTorch] Support lstm op for ExportedProgram importer [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Support lstm op for ExportedProgram importer [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Support lstm op for ExportedProgram importer [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Support lstm op for ExportedProgram importer [tvm] via GitHub
• (tvm) branch main updated: Fix conflict parameter name promote_dtye in FP8ComputeLegalize (#18334) tqchen
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@36e473f58bda75e03f745d30da09033d0ab880f9) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@36e473f58bda75e03f745d30da09033d0ab880f9) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@36e473f58bda75e03f745d30da09033d0ab880f9) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@36e473f58bda75e03f745d30da09033d0ab880f9) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@36e473f58bda75e03f745d30da09033d0ab880f9) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@36e473f58bda75e03f745d30da09033d0ab880f9) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@36e473f58bda75e03f745d30da09033d0ab880f9) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@36e473f58bda75e03f745d30da09033d0ab880f9) tqchen
• [PR] [FFI][ABI] Bump tvm-ffi version to reflect RC ABI Update [tvm] via GitHub
  • Re: [PR] [FFI][ABI] Bump tvm-ffi version to reflect RC ABI Update [tvm] via GitHub
  • Re: [PR] [FFI][ABI] Bump tvm-ffi version to reflect RC ABI Update [tvm] via GitHub
  • Re: [PR] [FFI][ABI] Bump tvm-ffi version to reflect RC ABI Update [tvm] via GitHub
• (tvm-ffi) branch main updated: [FFI][ABI] TVMFFIObject ABI to further align with torch intrusive_ptr (#57) tqchen
• (tvm) branch main updated: [Relax][PyTorch] Support MatrixMultiply op for ExportedProgram importer (#18343) ruihangl
• (tvm-ffi) branch main updated: fix: Type mismatch aborts on OpaquePyObject (#56) tqchen
• (tvm) branch main updated: [TIR] Support sequence comparisons in TVMScript (#18341) ruihangl
• (tvm-ffi) branch main updated: feat: Auto-create Python classes when missing (#49) tqchen
• (tvm) branch main updated: [Relax] Update BasePyModule with faster DLPack converter for tensor conversion (#18331) ruihangl
• (tvm-ffi) branch main updated: feat: Support `field(init=...)` in `@c_class` fields (#52) junrushao
• (tvm-ffi) branch main updated: fix: Compatibility with torch <= 2.8 (#54) junrushao
• [PR] [Relax][PyTorch] Support MatrixMultiply op for ExportedProgram importer [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Support MatrixMultiply op for ExportedProgram importer [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Support MatrixMultiply op for ExportedProgram importer [tvm] via GitHub
  • Re: [PR] [Relax][PyTorch] Support MatrixMultiply op for ExportedProgram importer [tvm] via GitHub
• [PR] Bump pip from 22.1.1 to 25.2 in /docker/python [tvm] via GitHub
  • Re: [PR] Bump pip from 22.1.1 to 25.2 in /docker/python [tvm] via GitHub
  • Re: [PR] Bump pip from 22.1.1 to 25.2 in /docker/python [tvm] via GitHub
• (tvm-site) branch asf-site updated: Build at Wed Sep 24 05:42:08 UTC 2025 github-bot
• (tvm) branch nightly updated (118e3b1316 -> a21e0df4b3) github-bot
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@a21e0df4b3e7a364d199f79d60994cc6154c8662) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@a21e0df4b3e7a364d199f79d60994cc6154c8662) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@a21e0df4b3e7a364d199f79d60994cc6154c8662) tqchen
• (tvm-ffi) branch main updated: chore: Revamp mypy config and publish typing metadata (#51) tqchen
• (tvm-site) branch asf-site updated: deploying docs (apache/tvm@e1f93f361ed80fe8407f7463be503bab656edf42) tqchen
  • (tvm-site) branch asf-site updated: deploying docs (apache/tvm@e1f93f361ed80fe8407f7463be503bab656edf42) tqchen
• (tvm) branch main updated: [Relax] Operator and RoPE support for Llama4 (#18336) ruihangl
• (tvm-ffi) branch main updated: feat: Support Array Concat in Python (#45) junrushao
• (tvm-ffi) branch main updated: chore: Add linters for `.rst`, `.md` and `.yaml` (#46) tqchen
• (tvm-ffi) branch main updated: feat: Improve Critical Path Perf in Cython `make_ret_object` (#47) tqchen
• [I] [Feature Request][Relax][Torch] Add support for torch.mm (mm.default) in from_exported_program [tvm] via GitHub
  • Re: [I] [Feature Request][Relax][Torch] Add support for torch.mm (mm.default) in from_exported_program [tvm] via GitHub
  • Re: [I] [Feature Request][Relax][Torch] Add support for torch.mm (mm.default) in from_exported_program [tvm] via GitHub
• [I] [Feature Request][Relax][Torch] Add support for nn.LSTM (lstm.input) in from_exported_program [tvm] via GitHub
  • Re: [I] [Feature Request][Relax][Torch] Add support for nn.LSTM (lstm.input) in from_exported_program [tvm] via GitHub
  • Re: [I] [Feature Request][Relax][Torch] Add support for nn.LSTM (lstm.input) in from_exported_program [tvm] via GitHub
  • Re: [I] [Feature Request][Relax][Torch] Add support for nn.LSTM (lstm.input) in from_exported_program [tvm] via GitHub
  • Re: [I] [Feature Request][Relax][Torch] Add support for nn.LSTM (lstm.input) in from_exported_program [tvm] via GitHub
• [I] [Bug][Relax][Torch] from_exported_program fails on unbind producing single-element tuple [tvm] via GitHub
  • Re: [I] [Bug][Relax][Torch] from_exported_program fails on unbind producing single-element tuple [tvm] via GitHub
  • Re: [I] [Bug][Relax][Torch] from_exported_program fails on unbind producing single-element tuple [tvm] via GitHub
• [PR] LLama 4 support [tvm] via GitHub
  • Re: [PR] LLama 4 support [tvm] via GitHub
  • Re: [PR] LLama 4 support [tvm] via GitHub
  • Re: [PR] LLama 4 support [tvm] via GitHub
  • Re: [PR] LLama 4 support [tvm] via GitHub
  • Re: [PR] [Relax] Operator and RoPE support for Llama4 [tvm] via GitHub
• [PR] Squashed llama4 [tvm] via GitHub
  • Re: [PR] Squashed llama4 [tvm] via GitHub
  • Re: [PR] Squashed llama4 [tvm] via GitHub
  • Re: [PR] Squashed llama4 [tvm] via GitHub
• [I] [Bug] conflict paramter name between docstring and definition [tvm] via GitHub
  • Re: [I] [Bug] conflict paramter name between docstring and definition [tvm] via GitHub
  • Re: [I] [Bug] conflict paramter name between docstring and definition [tvm] via GitHub
• (tvm) branch main updated: [Relax][Frontend][ONNX] Error converting operator Expand: TVMError: broadcast_to expects the input tensor shape is broadcastable to the target shape (#18329) tlopex
• (tvm-ffi) branch main updated: fix(c_class): Fix mypy warnings on `dataclasses.field(...)` (#43) tqchen
• (tvm-ffi) branch main updated: feat(typing): Parameterizable `Array[T]` and `Map[K, V]` (#37) tqchen
• (tvm-ffi) branch main updated: chore: Test Various Python Versions in GitHub Action CI (#41) junrushao
• (tvm-ffi) branch main updated: doc: Enable Sphinx Build Checks in CI (#38) junrushao
• (tvm-ffi) branch main updated: feat: Enable mypy type checking (#35) junrushao
• [PR] [FFI][ABI] Bump version ffi to latest [tvm] via GitHub
  • Re: [PR] [FFI][ABI] Bump version ffi to latest [tvm] via GitHub
• (tvm-ffi) branch main updated: [FIX] Fix device type override (#40) tqchen
• (tvm-ffi) branch dev deleted (was 46f7358) tqchen
• (tvm-site) branch asf-site updated: Build at Mon Sep 22 15:09:12 UTC 2025 github-bot
• (tvm-ffi) branch main updated: [REFACTOR][ABI] Update backtrace storage for append-friendly (#39) tqchen
• [PR] [Relax] Update BasePyModule with faster DLPack converter for tensor conversion [tvm] via GitHub
  • Re: [PR] [Relax] Update BasePyModule with faster DLPack converter for tensor conversion [tvm] via GitHub
  • Re: [PR] [Relax] Update BasePyModule with faster DLPack converter for tensor conversion [tvm] via GitHub
• (tvm) branch nightly updated (874de945ac -> 4c82c71933) github-bot
• (tvm) branch main updated: [flashinfer] Support directing JIT to FlashInfer GroupedGemm kernels (#18325) ruihangl
• (tvm) branch main updated: [Relax][Backend] Implement R.call_py_func operator for calling Python functions from compiled TVM (#18326) ruihangl
• (tvm-ffi) branch main updated: feat: Introduce Experimental `tvm_ffi.dataclasses.c_class` (#8) tqchen
• [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
  • Re: [PR] [Analyzer] Enhance ConstIntBoundAnalyzer and IntervalSet with modular set analysis [tvm] via GitHub
• (tvm-ffi) branch main updated: feat: Introduce `tvm_ffi::reflection::init<>` (#32) tqchen

• Earlier messages
• Later messages