fix(gemini): assign correct indices in batch embedding response

[lucassz]

Relevant issues

The Gemini batchEmbedContents response handler hardcoded index=0 for every embedding in the response. Any consumer relying on the OpenAI-format index field to match embeddings back to inputs would silently get wrong associations.

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/test_litellm/ directory, Adding at least 1 test is a hard requirement - see details
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem
• I have requested a Greptile review by commenting @greptileai and received a Confidence Score of at least 4/5 before requesting a maintainer review

Delays in PR merge?

If you're seeing a delay in your PR being merged, ping the LiteLLM Team on Slack (#pr-review).

CI (LiteLLM team)

CI status guideline:

• 50-55 passing tests: main is stable with minor issues.
• 45-49 passing tests: acceptable but needs attention
• <= 40 passing tests: unstable; be careful with your merges and assess the risk.

• Branch creation CI run
  Link:

• CI run for the last commit
  Link:

• Merge / cherry-pick CI run
  Links:

Screenshots / Proof of Fix

Added unit test asserting sequential indices and correct vector ordering for a batch response.

Type

🐛 Bug Fix

Changes

Use enumerate in process_response so each embedding gets its positional index instead of 0.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
litellm	Ready	Preview, Comment	Apr 13, 2026 11:05pm

[CLAassistant]

All committers have signed the CLA.

[lucassz]

@greptileai

[greptile-apps]

Greptile Summary

This PR fixes a bug in process_response within the Gemini batch embeddings transformation where every embedding in a batch response was assigned index=0 instead of its correct positional index. The one-line fix replaces the bare for embedding in loop with enumerate, and a new unit test verifies sequential indices and value ordering without making real network calls.

Confidence Score: 5/5

This PR is safe to merge — it's a minimal, well-tested bug fix with no side effects.

Single-line correctness fix with a targeted unit test that makes no real network calls. No backwards-incompatible changes, no security implications, and all custom rules are satisfied.

No files require special attention.

Important Files Changed

Filename	Overview
litellm/llms/vertex_ai/gemini_embeddings/batch_embed_content_transformation.py	Correct one-line fix: enumerate replaces hardcoded index=0, assigning proper sequential indices to batch embeddings.
tests/litellm/llms/vertex_ai/test_gemini_batch_embeddings.py	New unit test directly calls process_response with mock data, asserting correct sequential indices and value ordering; no real network calls.

_{Reviews (2): Last reviewed commit: "fix(gemini): assign correct indices in b..." | Re-trigger Greptile}

[codspeed-hq]

Merging this PR will not alter performance

✅ 16 untouched benchmarks

---

_{Comparing lucassz:fix/gemini-batch-embedding-indices (fbcf870) with main (548225e)}

[codecov]

Codecov Report

❌ Patch coverage is 0% with 1 line in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
...i_embeddings/batch_embed_content_transformation.py	0.00%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!