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

Merovingian May 21, 2024, 2:40pm 9

I encountered a similar problem last time, the outputs gave errors in almost all of them and I had to find the errors and solve them. I guess they will solve it with a micro-update

Topic		Replies	Views
Gpt-4o vs. gpt-4-turbo - function calling Bugs function-calling , gpt-4-turbo , gpt-4o	17	4967	September 23, 2024
Inconsistent and invalid JSON response API	8	5921	June 11, 2024
Structured Outputs not reliable with GPT-4o-mini and GPT-4o API structured-output	35	4151	December 21, 2024
Getting response data as a fixed & Consistent JSON response API	43	120795	February 19, 2024
Inconsistent Fine-Tuning Results: GPT-4o vs. GPT-4o-Mini API gpt-4	22	599	November 26, 2024

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

Related topics