Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
HF_TOKENNoHuggingFace token for private repos in model discovery
OLLAMA_URLNoOllama API endpointhttp://localhost:11434
EMBEDDER_MODENoEmbedding mode: 'local' uses Xenova transformer, 'api' forces OpenAI APIlocal
LM_STUDIO_URLNoLM Studio API endpointhttp://localhost:1234
OPENAI_API_KEYNoFallback API embedder key for semantic code search
VOYAGE_API_KEYNoAlternative API embedder key for semantic code search
LATENCY_ROUTINGNoEnables latency-based routing (set to 'true')
MCP_SERVERS_DIRNoDirectory for dynamic MCP server plugins./mcp-servers
LLM_GATEWAY_PORTNoHTTP server port3456
FALLBACK_STRATEGYNoEnables free-model fallback routing (e.g., 'free-models')
LOCAL_LLM_ENABLEDNoEnable local LLM routing (set to 'true')false
FREE_MODEL_CATALOGNoLoads the free-model catalog at startup (set to 'true')
LLM_GATEWAY_DB_PATHNoSQLite vault path~/.llm-gateway/vault.db
MCP_DYNAMIC_SERVERSNoEnable dynamic MCP server plugins (set to 'true')
AUTO_DISCOVER_MODELSNoEnable HF auto-discovery of local models at startupfalse
TRANSFORMERS_OFFLINENoSet to '1' to use only cached HuggingFace model, skip download
MODEL_ROUTING_ENABLEDNoEnable model routing (set to 'true')
LLM_GATEWAY_AUTH_TOKENNoBearer token for HTTP routes, must be at least 32 characters
LLM_GATEWAY_MASTER_KEYNo64-char hex key, otherwise auto-generated and saved to ~/.llm-gateway/master.key
MAX_COMPARISON_COST_USDNoCaps comparison-service spending
ENABLE_OUTPUT_COMPRESSIONNoEnable RTK-style output compressiontrue
LLM_GATEWAY_AUTH_REQUIREDNoForce auth on or off explicitly
OPTIMIZE_MESSAGES_ENABLEDNoEnable three-part prompt optimizationtrue
LLM_GATEWAY_SECURITY_PROFILENoSecurity profile for MCP tool exposure: local-dev, restricted, or openlocal-dev

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/JNZader/mcp-llm-bridge'

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