Sign in

v1

latestOpenAPI 3.1.0MIT2026-07-2614271305.9 KB

Qwen3.8-Max-Preview Responses API (OpenAI-Compatible)

🚧 This model is not yet available, stay tuned

  • Call Qwen3.8-Max-Preview using the OpenAI Responses protocol
  • Flexible input: input can be a string or a Chat-format message array (supporting multimodal input_text / input_image)
  • Multi-turn conversation: pass previous_response_id to link to the previous response, and the server automatically assembles the context (the response id is valid for 7 days)
  • Session caching: enable by adding x-dashscope-session-cache: enable to the request header (default disable); see usage.input_tokens_details.cached_tokens for hits
  • Thinking effort: controlled via reasoning.effort
  • Streaming output: when stream=true, returned as Responses events (response.output_text.delta / response.completed, etc.)
post/v1/responses

Headers

x-dashscope-session-cache'enable' | 'disable'

Session caching switch. When set to enable, the server automatically caches the conversation context to reduce multi-turn inference latency and cost.

Request body

model'qwen3.8-max-preview' required

Chat model name

instructionsstring

Inserted as a system instruction at the start of the context. When using previous_response_id, the previous turn's instructions are not carried into this turn.

previous_response_idstring

Unique ID of the previous response (the response id, valid for 7 days). Used to link multi-turn conversations; the server automatically retrieves and combines that turn's input and output as context.

max_output_tokensinteger

Maximum number of output tokens to generate this time (including thinking).

storeboolean

Whether to store this response.

  • true (default): can be referenced by previous_response_id
  • false: not stored and cannot be referenced later
streamboolean

Whether to return a Responses event stream.

temperaturenumber

Sampling temperature, range [0, 2].

top_pnumber

Nucleus sampling parameter, range (0, 1].

Example request

{
  "model": "qwen3.8-max-preview"
}

Response

Generated successfully

idstring

Unique ID of this response (valid for 7 days, can be used as previous_response_id)

objectstring
statusstring

Response status

modelstring
outputobject[]

Array of output items (including message / reasoning / built-in tool calls, etc.)

Example response

{
  "id": "resp_xxxxxxxx",
  "object": "response",
  "status": "completed",
  "model": "qwen3.8-max-preview",
  "usage": {
    "billing_rule": "per_call",
    "credits_reserved": 5,
    "user_group": "default"
  }
}