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.floor_divide on int64 lowers to a trapping sdiv, so a zero divisor raises SIGFPE instead of a defined result
- [Bug] [Performance Regression][Relax][CUDA] Fusing MaxPool with Add increases latency by 5.53x
- [Metal] Add TVM_METAL_STORAGE_MODE opt-in for correctness at scale (#20157)
- [Fix][CUDA] Preserve NaNs in floating-point min and max
- [Relax] Skip 1-D inputs in CombineParallelMatmul
- [Fix][Relax][Frontend][Torch] Promote the tensor, not the scalar, in binary ops with a Python scalar
- [Fix][Relax][Frontend][Torch] Follow torch's dtype rules for the division family
- [Fix][Relax][Frontend][Torch] Accumulate integer cumsum and cumprod in int64
- [Relax][Frontend][Torch] Add amax, amin and min.dim converters
- [Fix][Relax][Frontend][Torch] Make any and prod follow torch: missing overloads and result dtypes
- Docs
- Python not yet supported