Structured Output summary truncated
# support
j
Hello, i switched from the analysisPlan to artifactPlan ( summary/eval/structured data) -> structuredOutputs. the problem is that the summarys get gut for long calls because the structured output aparrently only supports about 513 characters. i increased the token limmit but that did not fix it. i need help as. soon as possible because this is impacting production calls.
example call id: 019c520d-e14e-755f-96b1-e90734a7767d
structured output config id: 8f83e29e-ec10-4fe9-aa89-f29b1d26ff25
s
this is typical issue with vapi structure issue
j
Hey, this sounds like a known limitation with structuredOutputs in Vapi AI where certain fields have a hard character cap that isn’t affected by token limits, so long-call summaries get truncated. In production, this usually happens when everything is pushed into one summary field. A common fix is splitting the output into multiple fields or moving long summaries back to artifact/analysis outputs and handling them in the webhook. I’ve helped teams resolve this by redesigning the schema and adjusting post-call processing. Quick question: are you currently using one large summary field, or multiple structured fields in your config? @Janko
j
one large summary field, and i want to keep one field. can i just safely continue to using analysis summary? because it says it will depricate soon. there must be a way to do summaries simply right? i am still using the webhook type processsing. i just switchted to taking the data from aratifacts.structured output instead of analysis... is that not how you should do it? because u said " going back to handling them in the webhook"?
c
Hey Janko, There's a field called minLength and maxLength in the structured output summary type. You can increase that from scheama setting to fix the truncating issue. https://cdn.discordapp.com/attachments/1471912071196184720/1473683440384610354/image.png?ex=69971a27&is=6995c8a7&hm=00e03bf1d0afef0577f3196765acf24e8f6caa329168c603ef15a031871a4ca9&