lightrag-client

Python SDK for the InfraX RAGservicer — hybrid knowledge-base (vector + graph + keyword) retrieval over REST.

Install

pip install lightrag-client

Quick Start

from lightrag_client import LightRAGClient

rs = LightRAGClient(base_url="http://localhost:9721", api_key="lr_xxx", tenant_id="market")

# Insert documents
rs.insert("market", "BTC 走势受美联储利率政策影响", "doc-1")
rs.insert_batch("market", [
    {"text": "DeFi TVL 回升", "doc_id": "doc-2"},
    {"text": "链上巨鲸增持", "doc_id": "doc-3"},
])

# Query (context retrieval, no LLM answer)
result = rs.query("market", "比特币走势")
print(result)

# Context-only retrieval with top_k
ctx = rs.retrieve("market", "利率", top_k=5)

# Documents
rs.list_documents("market", page=1, limit=20)
rs.delete("market", "doc-1")

Admin

# Tenants & API keys
rs.create_tenant("market", "Market Data", "default tenant")
rs.generate_api_key("market", "prod", expires_days=90)
rs.list_api_keys("market")

# Hot-reload LLM / Embedding config (no restart needed)
rs.update_config(llm={"api_key": "sk-new", "model": "deepseek-chat"})

All methods talk the standard InfraX envelope {code, message, data} and return the data payload on success; API failures raise LightRAGClientError with .status / .code / .message.

Requirements

• Python >= 3.9
• requests

License

MIT