> ## Documentation Index
> Fetch the complete documentation index at: https://docs.fact0.io/llms.txt
> Use this file to discover all available pages before exploring further.

# LlamaIndex

> Capture retrieval and LLM telemetry from LlamaIndex query engines.

# LlamaIndex

<Note>
  Native callback support is on the roadmap. The pattern below works today with any LlamaIndex query engine.
</Note>

## Pattern

Wrap query engine calls in a telemetry execution. Use `MODEL_INVOCATION` for the LLM call and `TOOL_CALL` for retrievers.

```python theme={null}
import fact0
from llama_index.core import VectorStoreIndex

client = fact0.Client(api_key="f0_live_...")
index = VectorStoreIndex.from_documents(documents)
query_engine = index.as_query_engine()

with client.telemetry.execution(agent_id="rag-bot") as ex:
    with ex.span("retrieve", span_type="TOOL_CALL") as span:
        nodes = query_engine.retrieve("What is Fact0?")
        span.complete(output={"hits": len(nodes)})

    with ex.span("synthesize", span_type="MODEL_INVOCATION") as span:
        response = query_engine.query("What is Fact0?")
        span.complete(output={"answer": str(response)})
```

## Coming soon

* `LlamaIndexCallbackHandler` for automatic instrumentation
* Per-node retrieval metadata (chunks, scores)

## Related

* [Telemetry client](/sdk/python/telemetry)
* [LangChain](/integrations/langchain)