Skip to content

Add FSDP CI and end-to-end FSDP tests + save/load fsdp#47357

Draft
3outeille wants to merge 13 commits into
split/a-pr-4-1-distributed-runtimefrom
split/a-pr-4-2-fsdp-tests-ci
Draft

Add FSDP CI and end-to-end FSDP tests + save/load fsdp#47357
3outeille wants to merge 13 commits into
split/a-pr-4-1-distributed-runtimefrom
split/a-pr-4-2-fsdp-tests-ci

Conversation

@3outeille

@3outeille 3outeille commented Jul 16, 2026

Copy link
Copy Markdown
Member

CI

Summary

Part of FSDP orchestration stack (2/2). Replaces #46990. #47352 (1/2) → this PR (2/2)

  • TODO: Add FSDP gather and DCP (distributed_checkpoint=True) paths in save_pretrained
  • Add FSDPTesterMixin and tests/test_fsdp_mixin.py (load/save/generation)
  • Add base_model_fsdp_plan to cohere2_moe (only model in this stack)
  • Add dedicated fsdp_ci CircleCI job and pytest markers

3outeille and others added 7 commits July 16, 2026 02:32
…SDP orchestration 2/3).

Route distributed loading and saving through DistributedMixin, migrate TP tests
and docs off tp_plan="auto", and add FSDP gather/DCP save paths.
Add FSDPTesterMixin, cohere2_moe base_fsdp_plan, dedicated fsdp_ci job, and
pytest markers for distributed FSDP load/save/generation coverage.
@3outeille 3outeille changed the title Wire DistributedConfig through from_pretrained and save_pretrained (F… Add FSDP CI and end-to-end FSDP tests + save/load fsdp Jul 16, 2026
@HuggingFaceDocBuilderDev

Copy link
Copy Markdown

The docs for this PR live here. All of your documentation changes will be reflected on that endpoint. The docs are available until 30 days after the last update.

@github-actions

Copy link
Copy Markdown
Contributor

[For maintainers] Suggested jobs to run (before merge)

run-slow: cohere2_moe

@github-actions

Copy link
Copy Markdown
Contributor

CI recap

Dashboard: View test results in Grafana
Latest run: 29477587778:1
Result: failure | Jobs: 14 | Tests: 142,585 | Failures: 90 | Duration: 11h 6m

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