This is the release note of v11.3.0. See here for the complete list of solved issues and merged PRs.

We are running a Gitter chat for general discussions and quick questions. Feel free to join the channel to talk with developers and users!

Highlights

Support for CUDA 11.8 & NVIDIA H100 GPUs

This release adds support for CUDA 11.8 and the latest NVIDIA H100 GPUs. Note that CUDA 11.8 support is included in the cupy-cuda11x wheel.

Support for Python 3.11

Wheels are now available for Python 3.11.

Changes

Enhancements

• Add wrapper for cutensorPermutation (#7083)
• Fix compile error from inf/nan in cupy.fuse (#7128)
• Support CUDA 11.8 (#7134)
• Add CUDA 11.8 on documents (#7148)
• Support NCCL 2.15 (#7160)
• Support CUDA 11.8 H100 GPUs (#7169)
• Support Python 3.11 (#7179)
• Fix indexing sparse matrix with empty index arguments (#7155)

Bug Fixes

• Make sure weibull distribution support ndarrays (#7055)
• Make sure that cupy (array-api) Array objects can be composed using asarray (#7095)
• JIT: Fix compile error for op.routine including in0_type (#7096)
• Don't use __del__ in TCPStore (#7111)
• Fix cupy.nansum in fusing (#7114)
• Fusion TypeError in cupy._core.fusion._call_ufunc() (#7130)
• JIT: Fix compile error of minmax function (#7174)

Documentation

• Docs: Add missing functions (#7112)

Installation

• Support force-overwriting Docker image via workflow (#7091)
• Bump version to v11.3.0 (#7182)

Tests

• CI: Add ROCm 5.3 (#7125)
• CI: Allow /test jenkins to trigger Jenkins only (#7129)
• Install zlib for CUDA 11.8 Windows CI (#7138)
• Fix for pytest 7.2 (#7149)
• CI: improve use of cache in GitHub Actions (#7156)
• CI: Add support for the latest FlexCI Windows image (#7172)

Others

• CI: use pre-commit in GitHub Actions (#7132)

Contributors

The CuPy Team would like to thank all those who contributed to this release!

@anaruse @andfoy @asi1024 @emcastillo @kmaehashi @leofang @takagi