| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| Parent folder | |||
| cutlass-install-x86_64-cu13-4.6.3.tar.gz.digest | 2026-08-27 | 107 Bytes | |
| cutlass-install-x86_64-cu13-4.6.3.tar.gz | 2026-08-27 | 117.7 MB | |
| cutlass-install-x86_64-cu12-4.6.3.tar.gz.digest | 2026-08-27 | 107 Bytes | |
| cutlass-install-aarch64-cu13-4.6.3.tar.gz.digest | 2026-08-27 | 108 Bytes | |
| cutlass-install-x86_64-cu12-4.6.3.tar.gz | 2026-08-27 | 118.2 MB | |
| cutlass-install-aarch64-cu13-4.6.3.tar.gz | 2026-08-27 | 119.7 MB | |
| cutlass-install-aarch64-cu12-4.6.3.tar.gz.digest | 2026-08-27 | 108 Bytes | |
| cutlass-install-aarch64-cu12-4.6.3.tar.gz | 2026-08-27 | 120.0 MB | |
| CUTLASS 4.6.3 source code.tar.gz | 2026-08-26 | 40.7 MB | |
| CUTLASS 4.6.3 source code.zip | 2026-08-26 | 50.1 MB | |
| README.md | 2026-08-26 | 1.8 kB | |
| Totals: 11 Items | 566.5 MB | 0 | |
CuTe DSL
- Bug fixing and improvements
- Fixed a failed kernel compilation issue when setmaxnreg is used together with specific warp-specialized patterns (!3382)
- Fixed a leak with jit/kernel decorator (!3421)
- Fixed bugs in cutlass.jax.cutlass_call with tensor aliasing and handling of optional tensors
- Reduced the protobuf version requirement of IKET profiler from 6.30 to 4.21
- Fixed import time write to CUTE_DSL_LIBS (!3462)
- Fixed export_to_c outputs i32 dynamic shapes in headers for i64 inputs (!3447)
- Improved performance of constructing
cutlass.Numericfrom Python value (!3443) - Fixed TVM-FFI env stream detection for GPU tensors in tuple (!3444)
- Fixed TVM-FFI error when passed a tensor with non-zero byte offset (!3450)
This release has been tested against the following packages:
- FlashAttention: main (a369df7)
- Quack: main (680ef82)
- FlashInfer: main (0263dc2)
- cuDNN-Frontend:deveop(ab9efe15)
- Pytorch: main(cf30153)
