Create chat completion (OpenAI-compatible)

Recent Requests
Log in to see full request history
TimeStatusUser Agent
Retrieving recent requests…
LoadingLoading…
Body Params
string
required

Model ID (e.g., qwen3.8-max, qwen3-coder-plus, qwen-deep-research)

messages
array of objects
required
messages*
boolean
Defaults to false
tools
array of objects
tools
web_search_options
object

OpenAI Chat Completions web-search options. Presence enables Qwen native web search.

boolean
deprecated

Legacy compatibility flag. true maps to thinking mode; false maps to fast mode. For qwen-deep-research, true selects advance research and false selects normal research. Prefer thinking_mode.

string
enum
Defaults to fast

Qwen thinking mode. fast disables reasoning; auto lets Qwen decide; thinking forces summarized reasoning. For qwen-deep-research, thinking selects advance research while fast and auto select normal research. Input handling is case-insensitive.

Allowed:
string
enum

OpenAI Chat Completions reasoning effort. For normal chat, none/minimal map to fast, low/medium map to auto, and high/xhigh/max map to thinking. For qwen-deep-research, medium/high/xhigh/max select advance research; omitted, none, minimal, or low select normal research.

Allowed:
Responses

Language
Credentials
Bearer
URL
LoadingLoading…
Response
Click Try It! to start a request and see the response here! Or choose an example:
application/json