Skip to main content
Glama
Heratiki

LocalLama MCP Server

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
LOG_LEVELNoLogging leveldebug
COST_THRESHOLDNoUSD cost above which local offload is preferred0.02
OLLAMA_TIMEOUTNoOllama timeout in seconds120
OLLAMA_ENDPOINTNoOllama API base URL
TOKEN_THRESHOLDNoToken count above which local offload is considered1500
QUALITY_THRESHOLDNoQuality score below which paid API is always used0.7
BENCHMARK_PARALLELNoRun benchmarks in parallelfalse
LLAMA_CPP_ENDPOINTNollama-server URL; leave unset to disable provider
LM_STUDIO_ENDPOINTNoLM Studio API base URL
LOCALLAMA_ROOT_DIRNoRoot directory for the server (overrides default project root)
OPENROUTER_API_KEYNoOpenRouter API key (enables OpenRouter provider)
CODE_SEARCH_ENABLEDNoEnable code searchtrue
DEFAULT_LOCAL_MODELNoModel name used when offloading to local provider
MIN_VALIDATOR_SCORENoMinimum validation score required before a model is eligible for external validation0.6
PROVIDER_TIMEOUT_MSNoProvider timeout in milliseconds120000
OPENROUTER_FREE_ONLYNoRestrict OpenRouter to free-tier models onlyfalse
BENCHMARK_RESULTS_PATHNoPath to save benchmark results./benchmark-results
BENCHMARK_SAVE_RESULTSNoSave benchmark resultstrue
BENCHMARK_TASK_TIMEOUTNoBenchmark task timeout in milliseconds60000
BENCHMARK_RUNS_PER_TASKNoNumber of benchmark runs per task3
REMOVE_STALE_LOCK_FILESNoAutomatically remove stale lock filestrue
VALIDATION_RETRY_BUDGETNoValidation retry attempts allowed after an initial failed validation1
RELIABLE_BENCHMARK_COUNTNoBenchmark runs required before empirical scores are treated as fully reliable3
CODE_SEARCH_INDEX_ON_STARTNoIndex code on startuptrue
EXPECT_LOCAL_PROVIDER_DOWNNoSet true in test to assert no local suggestion
BENCHMARK_MAX_PARALLEL_TASKSNoMaximum parallel benchmark tasks2
CODE_SEARCH_EXCLUDE_PATTERNSNoJSON array of glob patterns to exclude from indexing["node_modules/**","dist/**",".git/**"]
CODE_SEARCH_REINDEX_INTERVALNoReindex interval in seconds3600
PROVIDER_MAX_CONCURRENT_LOCALNoShared local execution slot count1
LOCK_FILE_CHECK_ACTIVE_PROCESSNoCheck for active process when lock file existstrue
PROVIDER_MAX_CONCURRENT_REMOTENoPer-remote-provider slot count5
PROVIDER_HEALTH_PROBE_INTERVAL_MSNoProvider health probe interval in milliseconds60000

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/Heratiki/locallama-mcp'

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