I could understand the default to 16 instead of infinity, but it should not break
{
"error": {
"message": "[{'type': 'int_type', 'loc': ('body', 'max_tokens'), 'msg': 'Input should be a valid integer', 'input': None, 'url': 'https://errors.pydantic.dev/2.4/v/int_type'}]",
"type": "internal_server_error",
"param": null,
"code": null
}
}
not sure if this is the api response or the error message that the client I'm using is logging, but the error is clear
If i set the max_token with a valid integer i get no errors
Calling the api with max_token null should work, but i get a 500 http response
per openai api doc https://platform.openai.com/docs/api-reference/chat/create#chat/create-max_tokens it should accept the null and fallback to default infinity
I could understand the default to 16 instead of infinity, but it should not break
not sure if this is the api response or the error message that the client I'm using is logging, but the error is clear If i set the max_token with a valid integer i get no errors