Issue Station
Sign In
Tools hereToolbox

vllm - ✅(Solved) Fix [Feature]: Sharded model loader doesn't support GCS [1 pull requests, 1 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
vllm-project/vllm#38568•Fetched 2026-04-08 01:53:20
View on GitHub
Comments
0
Participants
1
Timeline
2
Reactions
0
Author
kirilomankiriloman
Participants
kirilomankiriloman
Timeline (top)
cross-referenced ×1labeled ×1

Fix Action

Fixed

PR fix notes

PR #38557: [Feature] Add GCS support for sharded state loader

Description (problem / solution / changelog)

The ShardedStateLoader currently only supports S3 (via s3_glob) and local filesystem (via glob.glob) for discovering checkpoint files. GCS paths (gs://) fall through to glob.glob which cannot resolve remote URIs, causing runai_streamer_sharded to fail with GCS.

This adds a gcs_utils module mirroring s3_utils that uses google-cloud-storage (already a dependency via runai-model-streamer[gcs]) to list and filter blobs from GCS buckets. The sharded_state_loader is updated to use gcs_glob for gs:// paths.

Purpose

Support GCS in the runai_streamer_sharded load format.

Test Plan

Tested directly on a k8s cluster that pulls the model weights from GCS.

Test Result

Passed and the vLLM pod started-up successfully.

Closes https://github.com/vllm-project/vllm/issues/38568

Changed files

  • tests/model_executor/model_loader/test_gcs_utils.py (added, +102/-0)
  • vllm/model_executor/model_loader/sharded_state_loader.py (modified, +10/-2)
  • vllm/transformers_utils/gcs_utils.py (added, +97/-0)
RAW_BUFFERClick to expand / collapse

🚀 The feature, motivation and pitch

A PR has been opened to add that support: https://github.com/vllm-project/vllm/pull/38557.

Alternatives

No response

Additional context

No response

Before submitting a new issue...

  • Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.

extent analysis

Fix Plan

To address the issue, we will focus on reviewing and implementing the PR #38557.

Steps to Implement the Fix

  • Review the PR #38557 and ensure it covers all necessary changes.
  • Merge the PR into the main branch.
  • Update the documentation to reflect the new changes.

Example Code Changes

# Example of updated code
def new_function():
    # New functionality added in PR #38557
    pass

Verification

  • Verify the new functionality works as expected.
  • Run tests to ensure no regressions were introduced.

Extra Tips

  • Always review PRs thoroughly before merging.
  • Keep documentation up-to-date with code changes.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…
#embedding generation#cache error#pipeline error#runtime error#dependency conflict

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

vllm - ✅(Solved) Fix [Feature]: Sharded model loader doesn't support GCS [1 pull requests, 1 participants]