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

# On LLM Call • AI Agent SDK

> on_llm_call: Called when an LLM call is made.

# on\_llm\_call

<div className="flex items-center gap-2">
  <Badge color="purple">Method</Badge>
</div>

> This is a method of the [**ObservabilityAdapter**](../classes/ObservabilityAdapter) class in the [**obs**](../modules/obs) module.

Called when an LLM call is made.

## Signature

```python theme={"theme":{"light":"vitesse-light","dark":"vitesse-dark"}}
def on_llm_call(span_id: str, model: str, messages: List[Dict[str, Any]], response: Optional[str], tokens: Optional[Dict[str, int]], latency_ms: Optional[float]) -> None
```

## Parameters

<ParamField query="span_id" type="str" required={true}>
  No description available.
</ParamField>

<ParamField query="model" type="str" required={true}>
  No description available.
</ParamField>

<ParamField query="messages" type="List" required={true}>
  No description available.
</ParamField>

<ParamField query="response" type="Optional" required={false}>
  No description available.
</ParamField>

<ParamField query="tokens" type="Optional" required={false}>
  No description available.
</ParamField>

<ParamField query="latency_ms" type="Optional" required={false}>
  No description available.
</ParamField>

## Source

<Card title="View on GitHub" icon="github" href="https://github.com/MervinPraison/PraisonAI/blob/main/src/praisonai-agents/praisonaiagents/obs/__init__.py#L100">
  `praisonaiagents/obs/__init__.py` at line 100
</Card>

***

## Related Documentation

<CardGroup cols={2}>
  <Card title="Models Overview" icon="microchip" href="/docs/models" />

  <Card title="LLM Configuration" icon="gear" href="/docs/configuration/llm-config" />

  <Card title="Model Router" icon="route" href="/docs/features/model-router" />

  <Card title="Model Failover" icon="shield" href="/docs/features/failover" />
</CardGroup>