Text Generation (Responses)

POSThttps://api.tokenbyte.ai/v1/responses

Use Bearer authentication and Content-Type: application/json.

Request body

Field	Type	Required	Description
`model`	string	Yes	Full model ID supporting Responses
`input`	string/array	Yes	Plain text or structured input items
`instructions`	string	No	System-level instruction
`previous_response_id`	string	No	Continue from a stored response
`stream`, `store`	boolean	No	Streaming and persistence controls
`tools`, `tool_choice`	array/object	No	Tools and selection policy
`temperature`, `top_p`	number	No	Sampling controls
`reasoning`	object	No	Reasoning effort when supported

The output array can contain messages, reasoning, and tool calls. Read text from output_text content items rather than assuming the first output item is text. Streaming emits semantic events such as response.output_text.delta and response.completed.

curl https://api.tokenbyte.ai/v1/responses \
  -H "Authorization: Bearer $TOKENBYTE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"<model-id>","input":"Hello"}'

{
  "id": "resp_xxx",
  "object": "response",
  "status": "completed",
  "output": [
    {
      "type": "message",
      "role": "assistant",
      "content": [
        { "type": "output_text", "text": "Hello!", "annotations": [] }
      ]
    }
  ],
  "usage": { "input_tokens": 8, "output_tokens": 4, "total_tokens": 12 }
}