Skip to content

refactor(tests): migrate sharding_dump golden baselines from deepseek2-16b to deepseek3-test (Phase 1b-2) - #6367

Open
entrpn wants to merge 1 commit into
mainfrom
jfacevedo/model-deprecation-pr1b-sharding-goldens
Open

entrpn wants to merge 1 commit into
mainfrom
jfacevedo/model-deprecation-pr1b-sharding-goldens

Conversation

@entrpn

@entrpn entrpn commented Nov 16, 2026 •

Copy link
Copy Markdown
Collaborator

Description

Reviewer Note: Although GitHub shows 38 files changed, only 2 Python files (11 changed lines total) are hand-edited (tests/utils/sharding_dump.py and tests/utils/run_sharding_dump.py). All remaining 36 .json files under tests/utils/sharding_info/ are auto-generated sharding golden dumps produced by python3 -m tests.utils.run_sharding_dump.

This is PR 1b-2 of the phased MaxText Model Deprecation & Lifecycle Policy rollout (split out from #6320 so reviewers can inspect the 22 Python test fixture updates in #6320 separately from these auto-generated sharding JSON dumps):

  • PR 1a (feat(configs): add models/deprecated/ fallback resolution and Depreca… #6171, merged): Added configs/models/deprecated/<model_name>.yml fallback resolution and DeprecationWarning in pyconfig and pyconfig_deprecated.
  • PR 1b-1 (refactor(tests): migrate generic test fixtures off sunset models onto retained Tier 1/2 configs (Phase 1b-1) #6320 — independent companion PR): Migrates the 22 generic unit, integration, post-training, and MoE Python test files off sunset models onto retained Tier 1/2 configs.
  • PR 1b-2 (this PR): Migrates all 9 MoE sharding dump cases in tests/utils/sharding_dump.py off sunset deepseek2-16b onto retained Tier 1 deepseek3-test (and updates the CLI docstring example in tests/utils/run_sharding_dump.py from gemma-7b to gemma3-4b), removes the retired tests/utils/sharding_info/deepseek2-16b/ JSON goldens, and checks in freshly generated tests/utils/sharding_info/deepseek3-test/ (input_shardings.json, logical_shardings.json, and named_shardings.json) across all 9 topology/rule combinations (tpu7x-16, tpu7x-8, and v6e-16, including pure-fsdp, pipeline-large-moe, ep-as-cp, cp-as-ep, tp-as-ep, ep-as-dp, and shard-exp-on-fsdp).
  • PR 1c (next): Soft-sunset git mv of the sunset model configs into src/maxtext/configs/models/deprecated/ + docs/reference/models/tiering.md sync.
  • PR 1d: Automated Model Lifecycle Audit CLI (src/maxtext/tools/model_lifecycle_audit.py).

Tests

  • Regenerated and verified all 9 deepseek3-test sharding dump cases (27 JSON files across tpu7x-16, tpu7x-8, and v6e-16) against current main via python3 -m tests.utils.run_sharding_dump (libtpu AOT topology mesh compilation), leaving the 3 unrelated qwen3-0.6b / gpt-oss-20b baselines untouched.
  • Pre-commit linters (pyink --check, pylint, codespell): passed on tests/utils/sharding_dump.py and tests/utils/run_sharding_dump.py.

Checklist

Before submitting this PR, please make sure (put X in square brackets):

  • I have performed a self-review of my code. For an optional AI review, add the gemini-review label.
  • I have necessary comments in my code, particularly in hard-to-understand areas.
  • I have run end-to-end tests and provided workload links above if applicable.
  • I have made or will make corresponding changes to the doc if needed, including adding new documentation pages to the relevant Table of Contents (toctree directive) as explained in our documentation.

…2-16b to deepseek3-test (Phase 1b-2)

Migrate tests/utils/sharding_dump.py MoE golden test cases off sunset
deepseek2-16b onto retained Tier 1 deepseek3-test across all 9 TPU
topology/rule combinations, and regenerate the corresponding
tests/utils/sharding_info/ JSON golden files via run_sharding_dump.
@gemini-code-assist

Copy link
Copy Markdown

Summary of Changes

Hello, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request is part of the ongoing MaxText Model Deprecation & Lifecycle Policy initiative. The primary objective is to transition test infrastructure away from sunset models by migrating sharding dump test cases to supported configurations. This ensures that the sharding verification suite remains aligned with the current model support tier while maintaining test coverage.

Highlights

  • Model Migration: Migrated 9 MoE sharding dump test cases from the deprecated 'deepseek2-16b' model to the 'deepseek3-test' model.
  • Golden Dump Updates: Removed retired 'deepseek2-16b' JSON golden files and added newly generated 'deepseek3-test' sharding goldens across various topology and rule combinations.
  • CLI Documentation: Updated the CLI docstring example in 'tests/utils/run_sharding_dump.py' from 'gemma-7b' to 'gemma3-4b'.
New Features

🧠 You can now enable Memory (public preview) to help Gemini Code Assist learn from your team's feedback. This makes future code reviews more consistent and personalized to your project's style. Click here to enable Memory in your admin console.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature Command Description
Code Review /gemini review Performs a code review for the current pull request in its current state.
Pull Request Summary /gemini summary Provides a summary of the current pull request in its current state.
Comment @gemini-code-assist Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help /gemini help Displays a list of available commands.

Customization

To customize the Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counterproductive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for GitHub and other Google products, sign up here.

Footnotes

  1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

@github-actions github-actions Bot added the skip-ci-tests Add this label to the PR to skip running CI tests label Nov 16, 2026
@github-actions

Copy link
Copy Markdown
Contributor

🤖 Hi @entrpn, I'm running an automatic MaxText review of this pull request. You can track my progress in the logs. To request another review later, add the gemini-review label or comment @gemini-cli /review.

@github-actions github-actions Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

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

📋 Review Summary

Verdict: No significant issues found.

Inferred spec & validation
  • Sources: PR description, code after.
  • Problem: Sharding dump test baselines and CLI examples depend on models (deepseek2-16b, gemma-7b) that are being deprecated.
Spec item Implementation Validating test
R1 (stated): Migrate 9 MoE sharding dump cases to deepseek3-test. ✅ test_utils/sharding_dump.py ✅
R2 (stated): Update docstring example to gemma3-4b. ✅ test_utils/run_sharding_dump.py ✅
R3 (stated): Generate new JSON sharding golden files. ✅ test_utils/run_sharding_dump.py ✅

@codecov

codecov Bot commented Nov 16, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

skip-ci-tests Add this label to the PR to skip running CI tests

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants