Text Generation (Claude)

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

Send Authorization: Bearer <TOKENBYTE_API_KEY>, anthropic-version: 2023-06-01, and Content-Type: application/json.

Request body

Field	Type	Required	Description
`model`	string	Yes	Full model ID supporting Messages
`messages`	array	Yes	Alternating user and assistant messages
`max_tokens`	integer	Yes	Maximum output tokens
`system`	string/array	No	System instruction
`stream`	boolean	No	Return SSE events
`temperature`, `top_p`, `top_k`	number	No	Sampling controls
`stop_sequences`	array	No	Custom stop sequences
`tools`, `tool_choice`	array/object	No	Tools and selection policy
`thinking`	object	No	Extended thinking when supported

Message content may be a string or blocks such as text, image, tool_use, and tool_result. Non-streaming text is returned in content[].text; streaming uses Anthropic message and content-block events.

curl https://api.tokenbyte.ai/v1/messages \
  -H "Authorization: Bearer $TOKENBYTE_API_KEY" \
  -H "anthropic-version: 2023-06-01" \
  -H "Content-Type: application/json" \
  -d '{"model":"<model-id>","max_tokens":1024,"messages":[{"role":"user","content":"Hello"}]}'

{
  "id": "msg_xxx",
  "type": "message",
  "role": "assistant",
  "content": [{ "type": "text", "text": "Hello!" }],
  "stop_reason": "end_turn",
  "usage": { "input_tokens": 8, "output_tokens": 4 }
}