Skip to content

v25.12.16#1340

Merged
ROBERT-MCDOWELL merged 7 commits intomainfrom
v25
Dec 15, 2025
Merged

v25.12.16#1340
ROBERT-MCDOWELL merged 7 commits intomainfrom
v25

Conversation

@ROBERT-MCDOWELL
Copy link
Copy Markdown
Collaborator

No description provided.

Copilot AI review requested due to automatic review settings December 15, 2025 14:41
@ROBERT-MCDOWELL ROBERT-MCDOWELL merged commit 74b8a2a into main Dec 15, 2025
1 check passed
Copy link
Copy Markdown

Copilot AI left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Pull request overview

This PR (v25.12.16) includes changes to enhance GPU detection reliability and add compatibility workarounds for PyTorch's distributed module. The changes refactor GPU detection logic into reusable helper functions and adjust package installation flags.

Key Changes:

  • Added monkey-patching for PyTorch's distributed module to handle installations without distributed support
  • Refactored GPU detection logic into separate helper functions (has_working_cuda(), has_working_rocm(), has_working_xpu(), and corresponding PCI detection functions)
  • Removed --use-pep517 flag from bash script pip installation command

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 1 comment.

File Description
lib/classes/tts_engines/coqui.py Adds compatibility layer for PyTorch installations lacking torch.distributed module
lib/classes/device_installer.py Refactors GPU detection into modular helper functions and reorders detection priority (ROCm before CUDA)
ebook2audiobook.sh Removes --use-pep517 flag from pip install command in bash script

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

Comment thread lib/classes/tts_engines/coqui.py
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants