PyTorch 2.4.1 Release, bug fix release
AI Classified by Officially
This release is meant to fix the following issues (regressions / silent correctness): ### Breaking Changes: - The pytorch/pytorch docker image now installs the PyTorch package through pip and has switch its conda installation from miniconda to [miniforge](https://github.com/conda-forge/miniforge) ([#134274](https://github.com/pytorch/pytorch/pull/134274)) ### Windows: - Fix performance regression on Windows related to MKL static linking ([#130619](https://github.com/pytorch/pytorch/issues/130619)) ([#130697](https://github.com/pytorch/pytorch/pull/130697)) - Fix error during loading on Windows: [WinError 126] The specified module could not be found. ([#131662](https://github.com/pytorch/pytorch/issues/131662)) ([#130697](https://github.com/pytorch/pytorch/pull/130697)) ### MPS: - Fix tensor.clamp produces wrong values ([#130226](https://github.com/pytorch/pytorch/pull/130226)) - Fix Incorrect result from batch norm with sliced inputs ([#133610](https://github.com/pytorch/pytorch/pull/133610)) ### ROCM: - Fix for launching kernel invalid config error when calling embedding with large index ([#130994](https://github.com/pytorch/pytorch/pull/130994)) - Added a check and a warning w...
Read the original at the source: https://github.com/pytorch/pytorch/releases/tag/v2.4.1
Officially imported this from PyTorch’s own source. If you work there, claiming the profile and verifying the domain lets you choose to show the full text here.
This publication has changed since it was first published
2 versions recorded. The original is kept in full — nothing is overwritten.
- v2 imported change on current
- v1 as first published on
Provenance
- Organization
- PyTorch — imported from official source
- Official source
- https://api.github.com/repos/pytorch/pytorch/releases?per_page=25 API
- Imported
- September 15, 2026 19:08
- Versions
- 2 recorded
- Identity
173502660