Conversation State

Each text generation is independent and stateless.

You can manually handle mutli-turn conversations by passing additional messages as parameter to text generation request:

from openai import OpenAI

client = OpenAI()

response = client.responses.create(
  model="gpt-4o-mini",
  input=[
    {"role": "user", "content": "hello"},
    {"role": "assistant", "content": "hello"},
    {"role": "user", "content": "can you help me find a new job?"},
  ]
)

print(response.output_text)