Nyraxis AI
FeaturesHow it worksIntegrationsPricingDocs
Get started
← Integration guides

LlamaIndex Integration

Instrument your LlamaIndex RAG pipelines and agents with Nyraxis observability.

Install

pip install nyraxis-sdk

Quick start

from nyraxis_sdk import LlamaIndexHandler
from llama_index.core import Settings
from llama_index.core.callbacks import CallbackManager

handler = LlamaIndexHandler(
    api_key="nyx_your_api_key",
    base_url="http://localhost:8000",
    agent_name="my-rag-agent",
)

Settings.callback_manager = CallbackManager([handler])

# Run your query engine as normal
response = query_engine.query("What is AI governance?")

await handler.flush()

What gets captured

  • LLM calls — prompt, completion, token counts, model
  • Query events — query string, retrieved context, response
  • Embedding calls — model, input size
  • Cost auto-calculated from token counts
  • Governance policies evaluated (PII detection on outputs, cost limits)

Other integrations