Create Model Response
POST
/v1/responsesCreate a response using the OpenAI-compatible Responses API format.
Request Body
application/jsonmodel:requiredstringModel identifier used for this request.
Allowed
gpt-5.5Example
gpt-5.5input:requiredstringText, message, image, or file input supplied to the model.
Example
Tell me a three sentence bedtime story about a unicorn.instructions:requiredstringSystem or developer instructions added to the model context.
Allowed
You are a helpful assistent.max_output_tokens:requirednumberMaximum number of tokens allowed in the response.
Allowed
1024reasoning:requiredobjectReasoning configuration for supported models.
stream:requiredbooleanWhether to stream partial results as they become available.
temperature:requirednumberSampling temperature used to control response randomness.
Response
200A successful request returns a JSON response. The complete example is shown in the code panel.
Compatibility References
Use these external references to review additional parameters supported by compatible provider formats.