fix(gemini): add presence_penalty support for Google AI Studio

[Chesars]

Relevant issues

Fixes #14753

Pre-Submission checklist

• I have Added testing in the tests/litellm/ directory
• 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

Type

🐛 Bug Fix

Summary

presencePenalty` is a native Google Gemini API parameter as documented:

• Firebase GenerationConfig - presencePenalty?: number

When using presence_penalty with Google AI Studio Gemini models via LiteLLM, it throws an UnsupportedParamsError:

import litellm

response = litellm.completion(
    model='gemini/gemini-2.0-flash',
    messages=[{'role': 'user', 'content': 'Hello'}],
    presence_penalty=0.5
)
# ERROR: litellm.UnsupportedParamsError: gemini does not support parameters: ['presence_penalty']

Solution

Added presence_penalty to the list of supported OpenAI params in GoogleAIStudioGeminiConfig.get_supported_openai_params().

The parameter mapping to presencePenalty (camelCase) already existed in the parent class VertexGeminiConfig, so only the supported params list needed updating.

Tests Added

• test_google_ai_studio_presence_penalty_supported in tests/test_litellm/llms/vertex_ai/gemini/test_vertex_and_google_ai_studio_gemini.py

[vercel]

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

Project	Deployment	Review	Updated (UTC)
litellm	Ready	Preview, Comment	Dec 17, 2025 11:36pm