fix(test): update deprecated gemini embedding model

[shin-bot-litellm]

Test Fix

Failing Tests: test_gemini_embeddings

Root Cause: The text-embedding-004 model has been deprecated/removed from the Gemini API:

models/text-embedding-004 is not found for API version v1beta, or is not supported for embedContent

Fix: Update the test to use gemini/gemini-embedding-001 which is the current supported embedding model for Gemini.

Changed Files:

• tests/local_testing/test_embedding.py - Replace gemini/text-embedding-004 → gemini/gemini-embedding-001

[vercel]

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

Project	Deployment	Actions	Updated (UTC)
litellm	Ready	Preview, Comment	Feb 7, 2026 2:02am

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.

---

Shin seems not to be a GitHub user. You need a GitHub account to be able to sign the CLA. If you have already a GitHub account, please add the email address used for this commit to your account.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[greptile-apps]

Greptile Overview

Greptile Summary

Updated the test_gemini_embeddings test to use gemini/gemini-embedding-001 instead of the deprecated gemini/text-embedding-004 model, which was failing with a "not found for API version v1beta" error.

• Changed model identifier from gemini/text-embedding-004 → gemini/gemini-embedding-001 in both sync and async test paths
• The new model is already properly configured in the model pricing configuration file
• This is a straightforward test fix with no logic changes

Confidence Score: 5/5

• This PR is safe to merge with minimal risk
• The change is a simple model identifier update in a test file, replacing a deprecated model with a supported one. The new model is already used elsewhere in the codebase and properly configured in the model pricing file. No logic changes, no security concerns, and directly addresses the failing test.
• No files require special attention

Important Files Changed

Filename	Overview
tests/local_testing/test_embedding.py	Updated test to use gemini/gemini-embedding-001 instead of deprecated gemini/text-embedding-004 model

Sequence Diagram

sequenceDiagram
    participant Test as test_gemini_embeddings
    participant LiteLLM as litellm.embedding/aembedding
    participant Gemini as Gemini API
    
    Test->>Test: Set verbose mode
    alt Sync Mode
        Test->>LiteLLM: embedding(model="gemini/gemini-embedding-001", input)
        LiteLLM->>Gemini: Request embedding with new model
        Gemini-->>LiteLLM: Return embedding response
        LiteLLM-->>Test: Return response object
    else Async Mode
        Test->>LiteLLM: aembedding(model="gemini/gemini-embedding-001", input)
        LiteLLM->>Gemini: Request embedding with new model (async)
        Gemini-->>LiteLLM: Return embedding response
        LiteLLM-->>Test: Return response object
    end
    Test->>Test: Assert embedding is list
    Test->>Test: Assert usage.prompt_tokens > 0

Loading

[greptile-apps]

_{1 file reviewed, no comments}

_{Edit Code Review Agent Settings | Greptile}