Hi, I have a call that is 11 minutes, where the successEvaluation indicates a perfect success, however it did not give any structuredData. In the logs I can see that it got an error from all models when trying to create the structuredData. Other calls create the structuredData just fine. Is this something that I did wrong on my end or was it an error on VAPIs side? Also, is it possible to retry or rerrun the analysisPlan somehow? Or do i have to do my own analysis on my own backend if the VAPI one failed? I attach some screenshots (we are mainly using VAPI api):

KEY
VALUE
time
Sep 15, 17:36:18.553
level
ERROR
body
Google GenerateContent error
attributes.category
model
attributes.callId
9ff7130e-f720-45e9-92f8-889c572f7046
attributes.orgId
f3a148cc-7c14-4337-917a-51ea93cfde61
attributes.provider
google
attributes.model
gemini-2.5-flash
attributes.error.message
Google API request timed out after 5000ms. model = gemini-2.5-flash
attributes.error.name
Error
attributes.error.stack
Error: Google API request timed out after 5000ms. model = gemini-2.5-flash at Timeout._onTimeout (/app/dist/libs/core/src/thirdPartyFunctions/googleFunctions.js:304:28) at listOnTimeout (node:internal/timers:581:17) at process.processTimers (node:internal/timers:519:7)
attributes.metrics.latencyMs
5001

https://cdn.discordapp.com/attachments/1417489705976467509/1417489706429448272/Screenshot_2025-09-16_at_14.36.25.png?ex=68caaba9&is=68c95a29&hm=88269fe1e0d578b1faab05f62039cac4680aa0b276224e67d95b4f7fded7861a& https://cdn.discordapp.com/attachments/1417489705976467509/1417489707117445130/Screenshot_2025-09-16_at_14.36.17.png?ex=68caaba9&is=68c95a29&hm=8b0f33f1b515c6118a68f806653738d988a4efa82420c1b4fc6bdb40a317113a&

Based on the documentation, VAPI does not currently offer a built-in way to retry or rerun the analysisPlan for a call after it has failed. If structuredData extraction fails due to model errors, your best option is to retrieve the call transcript via the API and perform your own analysis on your backend. Example:

bash
curl -H "Authorization: Bearer <token>" https://api.vapi.ai/call/<callId>

This will let you access the transcript for custom processing. [More info](https://docs.vapi.ai/assistants/call-analysis) Source: - [Call analysis documentation](https://docs.vapi.ai/assistants/call-analysis)

okay thanks