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

# Troubleshoot high latency

> Follow these steps to resolve high end-to-end latency issues.

If your end-to-end latency P90 exceeds **3 seconds**, follow these troubleshooting steps.

<Steps>
  <Step title="Check Estimated Latency">
    1. Look for features marked with a turtle icon 🐢 in your configuration
    2. Check if your estimated latency is higher than 1.5s
    3. If high, disable some features to reduce latency

    <Frame>
      <img height="700" src="https://mintcdn.com/retellai/M9QYKZE4hbt00HfL/images/l_2.png?fit=max&auto=format&n=M9QYKZE4hbt00HfL&q=85&s=e905fc5c601e1818911cce8ba65ec0fe" alt="Features that increase latency" data-path="images/l_2.png" />
    </Frame>
  </Step>

  <Step title="Monitor LLM Response Time">
    1. Check your current LLM latency
    2. Compare against normal range (500ms - 900ms)
    3. If consistently above 900ms:
       * Use [fast tier](/build/llm-options#fast-tier)
       * Switch to an alternative LLM provider
       * Or wait for your current provider to resolve performance issues
  </Step>

  <Step title="Check Status Page">
    1. Check the [Status Page](https://status.retellai.com/) for any ongoing issues
    2. If there are any ongoing issues, wait for them to resolve
  </Step>

  <Step title="Consider Geographic Distance">
    International phone calls may introduce additional latency due to geographic distance:

    1. Check if you're making calls between different countries or continents
    2. Consider getting a local phone number in the same region as your users
    3. Use phone numbers geographically closer to your target audience for better performance
  </Step>

  <Step title="Contact Support">
    If the above steps don't resolve your latency issues:

    1. Locate your call ID
    2. Message the [support](/general/support)
    3. Include:
       * Your call ID
       * Steps you've already tried
       * Current latency measurements
  </Step>
</Steps>