I’m using {“tool_choice”: “required”} to guarantee a Pydantic structured response. HOWEVER, I noticed 4o does not consistently respect the JSON schema. For example, if it chooses to return markdown, it will completely ignore the required json schema. GPT-4 Turbo on the other hand does this very con…

GPT-4o doesn't consistently respect JSON schema on tool use

snlamb May 15, 2024, 7:20am 2

Yes, I just experienced that as well - where gpt-4o does not respect the JSON mode requirement.

Topic		Replies	Views
GPT-4o API giving wild responses Feedback gpt-4o	8	3896	June 10, 2024
Function JSON Schema is still ignored by GPT-4, 4o, and 4o-mini when calling tools Bugs api	3	557	August 6, 2024
Gpt-4o vs. gpt-4-turbo - function calling Bugs function-calling , gpt-4-turbo , gpt-4o	17	6033	September 23, 2024
Json format causes infinite "\n \n \n \n" in response API gpt-4 , api , json-mode	21	9984	April 30, 2025
Getting response data as a fixed & Consistent JSON response API	43	128546	February 19, 2024