[link89]

Summary

• restrict pin_pytorch_gpu to torch>=2.11.0,<2.12.0
• avoid dp compress failing because torch.compile does not support the previously pinned PyTorch 2.10
• the CPU dependency group was already pinned to PyTorch 2.11.0

Error

deepmd torch.compile support requires PyTorch 2.11.x, 2.12.x, 2.13.x; found torch 2.10.

Summary by CodeRabbit

• Chores
  • No user-facing changes are noted in this update. The release summary identifies no changes to the app’s visible features, interface, or workflows. Existing user-facing behavior is not described as changing, and no new features or fixes are listed. These notes therefore focus on the absence of reported end-user impact rather than internal implementation details.

[coderabbitai]

No actionable comments were generated in the recent review.

Walkthrough

The pin_pytorch_gpu dependency changes from exactly Torch 2.10.0 to versions >=2.11.0 and <2.12.0.

Changes: GPU Torch Dependency

Layer / File(s)	Summary
Update GPU Torch version constraint pyproject.toml	pin_pytorch_gpu now accepts Torch versions from 2.11.0 inclusive to 2.12.0 exclusive.

Estimated code review effort: 1 (Trivial) | ~5 minutes

Change: Bug fix

Possibly related issues

• deepmodeling/deepmd-kit issue 5480 reports instability with PyTorch 2.10 in SeZM torch.compile and AOT-export paths, and calls for CI to use a compatible PyTorch version.

Possibly related PRs

• deepmodeling/deepmd-kit#6085: Updates the PyTorch Exportable compression workflow, which is also connected to the reported dp compress failure.
• deepmodeling/deepmd-kit#6046: Documents supported PyTorch versions for DPA4/SeZM torch.compile, including the version range selected here.
• deepmodeling/deepmd-kit#6110: Changes compiled DPA4C inference and Inductor compile options in the GPU torch.compile path.

Suggested labels: Python

Suggested reviewers: wanghan-iapcm

Merge Risk: Minimal · up to 88b66

This change moves the GPU dependency to the PyTorch 2.11 line that the compression path requires, so it should fix the reported dp compress failure. No concrete merge-blocking problem was established. Users on older GPUs or drivers who install from the default PyPI index may need to pick a CUDA-specific PyTorch build themselves.

Pre-merge checks: 5

Check name	Status	Explanation
Description Check		Check skipped - CodeRabbit’s high-level summary is enabled.
Title check		The title clearly summarizes the main change: updating the GPU dependency to require PyTorch 2.11.
Docstring Coverage		No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentat…
Linked Issues check		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.

Full details: Docstring Coverage

Explanation

No eligible functions found in the changed files. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff with cyclomatic complexity of at least 2. Documentation-only edits and deletions within surviving functions are included; unchanged diff context and fully deleted functions are excluded. Analyzed 0 functions across 0 files. (1 skipped: 1 unsupported.)

Finishing Touches

Generate unit tests (beta)

• Create a new PR

• Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

Share

• X
• Mastodon
• Reddit
• LinkedIn

Recent review info

Run configuration

• Source mode: Auto
• Repository: Unavailable
• Branch: Unavailable
• File: No configuration file from this repository applied
• Commit: Unavailable
• Layers: Repository UI
• Review profile: CHILL
• Plan: Advanced
• Run ID: 9f590135-29b7-4154-926d-1129250f223d

Commits

Reviewing files that changed from the base of the PR and between 3b2555c and 88b66d7.

Files selected for processing: 1

• pyproject.toml

Included review availability: This review used your included allowance. Your plan provides up to 8 included reviews per hour; 7 remain after this review.