tvm
https://github.com/apache/tvm
Python
Open deep learning compiler stack for cpu, gpu and specialized accelerators
Triage Issues!
When you volunteer to triage issues, you'll receive an email each day with a link to an open issue that needs help in this project. You'll also receive instructions on how to triage issues.
Triage Docs!
Receive a documented method or class from your favorite GitHub repos in your inbox every day. If you're really pro, receive undocumented methods or classes and supercharge your commit history.
Python not yet supported3 Subscribers
Add a CodeTriage badge to tvm
Help out
- Issues
- [Bug][Relax] ONNX Slice with runtime negative starts gives wrong results (zeros) — dynamic_strided_slice mishandles negative begin
- [Fix][Relax] Validate non-concat axes before fusing parallel matmul biases
- [Bug][Relax] CPU default pipeline: fused PrimFunc keeps symbolic height/width unbound (MakePackedAPI ICHECK)
- [Relax][VM] Re-running KillAfterLastUse on a default-lowered module crashes compiled VM codegen[Bug]
- [Bug][Relax] CombineParallelMatmul fails valid branches with broadcast-compatible biases
- [Bug][Relax][Frontend][ONNX] Softmax ignores opset<=12 coerce-to-2D semantics and defaults `axis` to -1 instead of 1, silently producing wrong output for valid models
- [Bug] [Bug][Relax] StaticPlanBlockMemory can emit non-dominating storage across if branches for reshape chains
- [Bug][Relax] Default LLVM pipeline can leave high-level Relax ops after FoldConstant resolves dynamic reshape shape
- [Bug][Relax] Default LLVM pipeline crashes in FoldConstant when reshape target is a runtime Shape
- [Bug][Relax] FuseTIR crashes when FuseOps creates a private function named fused for nested tuple getitem[Bug]
- Docs
- Python not yet supported