Skip to main content
Glama

Server Configuration

Describes the environment variables required to run the server.

NameRequiredDescriptionDefault
RAG_DEVICENoDevice for inference: cpu or cudacpu
RAG_OFFLINENoWhether to run in offline mode (true/false)false
RAG_HF_TOKENNoOptional Hugging Face access token for model downloads
RAG_DATA_PATHNoBase path for data storagedata
RAG_LOG_LEVELNoLogging levelINFO
RAG_NLI_MODELNoHugging Face model name used for natural language inferenceMoritzLaurer/mDeBERTa-v3-base-mnli-xnli
RAG_CACHE_PATHNoPath where local models are cacheddata/cache
RAG_CHUNK_SIZENoMaximum chunk size for extracted text512
RAG_QDRANT_URLNoURL of the Qdrant serverhttp://127.0.0.1:6333
RAG_RESULT_COUNTNoNumber of final search results returned5
RAG_CHUNK_OVERLAPNoOverlap between consecutive chunks64
RAG_METADATA_PATHNoPath for metadata and index manifestsdata/metadata
RAG_DOCUMENTS_PATHNoPath to the directory containing PDF documentsdocuments
RAG_RERANKER_MODELNoHugging Face model name used for rerankingBAAI/bge-reranker-v2-m3
RAG_EMBEDDING_MODELNoHugging Face model name used for embeddingsBAAI/bge-m3
RAG_SOURCE_DIVERSITYNoWhether to diversify the sources of returned passages (true/false)true
RAG_QDRANT_COLLECTIONNoName of the Qdrant collectionacademic_chunks
RAG_VECTOR_CANDIDATESNoNumber of candidate vectors retrieved from Qdrant before reranking20
RAG_VERIFICATION_PATHNoPath for verification and revision reportsdata/verification
RAG_EMBEDDING_BATCH_SIZENoBatch size for embedding ingestion16
RAG_ENTAILMENT_THRESHOLDNoConfidence threshold for an entailment (SUPPORTED) verdict0.7
RAG_CONTRADICTION_THRESHOLDNoConfidence threshold for a contradiction (CONTRADICTED) verdict0.7
RAG_DEDUPLICATION_THRESHOLDNoSimilarity threshold for deduplicating overlapping passages0.9

Instructions

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Capabilities

Features and capabilities supported by this server

Protocol revision2025-11-25

CapabilityDetails
tools
{
  "listChanged": false
}
prompts
{
  "listChanged": false
}
resources
{
  "subscribe": false,
  "listChanged": false
}
experimental
{}

Tools

Functions exposed to the LLM to take actions

NameDescription
search_referencesA

Search local indexed academic passages. Inspect the exact text before citing it: relevance is not evidence of entailment, and missing source metadata must never be invented.

get_referenceA

Resolve one citation ID to its exact stored passage and known provenance. Do not infer or invent fields absent from the result.

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/ehsanghorbani190/thegenie'

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