NVIDIA/TileGym v1.1.0

v1.1.0

Repository: NVIDIA/TileGym

Tag: v1.1.0

Published: 2026-04-03T10:27:40Z

Prerelease: no

Release notes:

What's Changed

• Igore vuln in pygments by @xjmxyt in https://github.com/NVIDIA/TileGym/pull/86
• Fix random mhc test and benchmark failures & Add unsloth geglu and grouped_gemm kernels by @hannahli-nv in https://github.com/NVIDIA/TileGym/pull/84
• bench: replace unfused reference_rms_norm with F.rms_norm as PyTorch baseline by @xjmxyt in https://github.com/NVIDIA/TileGym/pull/89
• Add tileiras optional dependency for bundled compiler support by @hannahli-nv in https://github.com/NVIDIA/TileGym/pull/90
• Add sparse MLA forward op in experimental by @Weili-0234 in https://github.com/NVIDIA/TileGym/pull/91
• Bump CUDA base image from 13.1.0 to 13.2.0 by @hannahli-nv in https://github.com/NVIDIA/TileGym/pull/85
• Bump version from 1.0.1 to 1.1.0 by @hannahli-nv in https://github.com/NVIDIA/TileGym/pull/94
• Remove cuda-tile-experimental URL dep for PyPI compatibility by @hannahli-nv in https://github.com/NVIDIA/TileGym/pull/95

Full Changelog: https://github.com/NVIDIA/TileGym/compare/v1.0.1...v1.1.0