Create chat completion

POST

/chat/completions

curl --request POST \
  --url https://api.mixen.ai/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '{ "messages": [ { "content": "Hello!", "role": "user" } ], "model": "gpt-5.5", "stream": false }'

OpenAI-compatible chat completion. Set stream=true for SSE token streaming; otherwise returns a single chat.completion object. Charged per-token from balance. Supports vision (image_url data: parts on vision models) and Perplexity Sonar (category=search models) — web results return as message.annotations (url_citation).

Authorizations

• ApiKeyAuth

Request Body^required

Media typeapplication/json

ChatCompletionRequest

object

model

required

Model

Model id from GET /models, or a common OpenAI/Claude/Gemini alias.

string

messages

required

Messages

Conversation so far; the last user message is the prompt.

Array<object>

_Msg

object

role

required

Role

string

content

Content

tool_calls

Any of:

Array<object>

Array<object>

object

key

additional properties

any

null

null

tool_call_id

Any of:

string

string

null

null

stream

Stream

If true, stream tokens as SSE (text/event-stream).

boolean

stream_options

Any of:

object

object

key

additional properties

any

null

null

temperature

Any of:

number

number

null

null

max_tokens

Any of:

integer

integer

null

null

max_completion_tokens

Any of:

integer

integer

null

null

reasoning_effort

Any of:

string

string

Allowed values: off low medium high xhigh max

null

null

tools

Any of:

Array<object>

Array<object>

object

key

additional properties

any

null

null

tool_choice

Any of:

Tool Choice

Tool Choice

null

null

response_format

Any of:

object

object

key

additional properties

any

null

null

web_search

Any of:

boolean

boolean

string

string

Allowed values: auto native exa

null

null

provider

Any of:

object

object

key

additional properties

any

null

null

service_tier

Any of:

string

string

Allowed values: flex priority

null

null

stop

Any of:

Array<string>

Array<string>

<= 4 items

null

null

seed

Any of:

integer

integer

null

null

top_p

Any of:

number

number

<= 1

null

null

top_k

Any of:

integer

integer

null

null

min_p

Any of:

number

number

<= 1

null

null

frequency_penalty

Any of:

number

number

>= -2 <= 2

null

null

presence_penalty

Any of:

number

number

>= -2 <= 2

null

null

repetition_penalty

Any of:

number

number

<= 2

null

null

logprobs

Any of:

boolean

boolean

null

null

top_logprobs

Any of:

integer

integer

<= 20

null

null

plugins

Any of:

Array<object>

Array<object>

object

key

additional properties

any

null

null

Example

{
  "messages": [
    {
      "content": "Hello!",
      "role": "user"
    }
  ],
  "model": "gpt-5.5",
  "stream": false
}

Responses

200

Successful Response

Media typeapplication/json

Example^generated

example

422

Validation Error

Media typeapplication/json

HTTPValidationError

object

detail

Detail

Array<object>

ValidationError

object

loc

required

Location

Array

msg

required

Message

string

type

required

Error Type

string

input

Input

ctx

Context

object

Example^generated

{
  "detail": [
    {
      "loc": [
        "example"
      ],
      "msg": "example",
      "type": "example",
      "input": "example",
      "ctx": {}
    }
  ]
}