Skip to main content
Glama
mnthe
by mnthe

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
GEMINI_MODELNoModel to use (default: gemini-3.5-flash).
GEMINI_TOP_KNoTop-k sampling parameter (default: 40).
GEMINI_TOP_PNoTop-p sampling parameter (default: 0.95).
GEMINI_API_KEYNoYour Google AI Studio API key. Required for AI Studio mode.
GEMINI_LOG_DIRNoDirectory for log files (default: ./logs).
GEMINI_MAX_TOKENSNoMaximum output tokens (default: 8192).
GEMINI_MAX_HISTORYNoMaximum history entries per session (default: 10).
GEMINI_MCP_SERVERSNoJSON array of external MCP server configurations.
GEMINI_TEMPERATURENoTemperature for generation (default: 1.0).
GEMINI_LOG_TO_STDERRNoLog to stderr instead of files (default: true).
GEMINI_SYSTEM_PROMPTNoCustom system prompt to set the AI's persona.
GOOGLE_CLOUD_PROJECTNoYour GCP project ID. Required for Vertex AI mode.
GOOGLE_CLOUD_LOCATIONNoGCP location (e.g., us-central1). Required for Vertex AI mode.
GEMINI_ALLOW_FILE_URISNoAllow file:// URIs (default: false, secure).
GEMINI_DISABLE_LOGGINGNoDisable logging entirely (default: false).
GEMINI_SESSION_TIMEOUTNoSession timeout in seconds (default: 3600).
GEMINI_IMAGE_OUTPUT_DIRNoDirectory for generated images (default: ~/Pictures/gemini-generated).
GEMINI_MEDIA_RESOLUTIONNoMedia resolution for Gemini 3 models: low, medium, high.
GEMINI_MUSIC_OUTPUT_DIRNoDirectory for generated music (default: ~/Music/gemini-generated/music).
GEMINI_VIDEO_OUTPUT_DIRNoDirectory for generated videos (default: ~/Movies/gemini-generated on macOS, ~/Videos/gemini-generated on Windows/Linux).
GEMINI_SPEECH_OUTPUT_DIRNoDirectory for generated speech (default: ~/Music/gemini-generated/speech).
GOOGLE_GENAI_USE_VERTEXAINoSet to 'false' to use Google AI Studio mode. Default is 'true' for Vertex AI mode.
GEMINI_ENABLE_CONVERSATIONSNoEnable multi-turn conversations (default: false).
GOOGLE_APPLICATION_CREDENTIALSNoPath to service account key JSON file for Vertex AI authentication (alternative to ADC).

Capabilities

Server capabilities have not been inspected yet.

Tools

Functions exposed to the LLM to take actions

NameDescription

No tools

Prompts

Interactive templates invoked by user choice

NameDescription

No prompts

Resources

Contextual data attached and managed by the client

NameDescription

No resources

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/mnthe/gemini-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server