Nyraxis AI
FeaturesHow it worksIntegrationsPricingDocs
← Integration guides

LlamaIndex Integration

Full observability for your LlamaIndex RAG pipelines — zero code changes.

Install

pip install nyraxis-sdk

Quick start

import nyraxis_sdk
from llama_index.core import VectorStoreIndex, SimpleDirectoryReader

nyraxis_sdk.init(
    api_key="nyx_your_api_key",
    agent_name="my-rag-agent",
)

# Use LlamaIndex as normal — all calls are auto-traced
documents = SimpleDirectoryReader("data").load_data()
index = VectorStoreIndex.from_documents(documents)
query_engine = index.as_query_engine()
response = query_engine.query("What is AI governance?")

nyraxis_sdk.shutdown()

What gets captured

  • LLM calls — prompt, completion, token counts, model, cost
  • Embedding calls — model, input size
  • Retrieval — query, retrieved context
  • Governance policies evaluated (PII detection, cost limits)

Other integrations