Skip to main content
POST
/
v1
/
chat
/
completions
curl --request POST \
  --url https://api.powertokens.ai/v1/chat/completions \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "qwen3-max",
  "messages": [
    {
      "role": "system",
      "content": "You are a helpful assistant."
    },
    {
      "role": "user",
      "content": "Hello, please introduce yourself."
    }
  ],
  "temperature": 0.7,
  "max_tokens": 2000
}
'
{
  "id": "<string>",
  "object": "chat.completion",
  "created": 123,
  "model": "<string>",
  "choices": [
    {
      "index": 123,
      "message": {
        "content": "<string>",
        "tool_calls": [
          {}
        ]
      },
      "finish_reason": "<string>"
    }
  ],
  "usage": {
    "prompt_tokens": 123,
    "completion_tokens": 123,
    "total_tokens": 123
  }
}

Authorizations

​
Authorization
string
header
required

Pass Authorization: Bearer <token> in the request header.

Body

application/json
​
model
enum<string>
required

Model name. Only publicly available chat models in the current Ali channel's final whitelist are supported.

Available options:
qwen3-max
Example:

"qwen3-max"

​
messages
object[]
required

List of conversation messages. Supports text content and multimodal content (image + text).

​
stream
boolean
default:true

Whether to use streaming output. Defaults to false.

​
temperature
number
default:0.7

Sampling temperature, controls output randomness. Range [0, 2].

Required range: 0 <= x <= 2
​
top_p
number
default:0.95

Nucleus sampling probability threshold.

Required range: 0 <= x <= 1
​
max_tokens
integer

Maximum number of tokens to generate.

Required range: x >= 1
​
stop
string[]

List of stop sequences.

​
seed
integer

Random seed for reproducible generation results.

​
tools
object[]

List of tools for Function Calling.

​
tool_choice
string

Tool selection strategy. Possible values: auto, none, required.

​
enable_search
boolean

Whether to enable search-augmented generation. Only supported by some models.

​
response_format
object

Response format, used to force the model to output JSON.

Response

Success. Returns the chat completion result. In streaming mode, returns an SSE stream.

​
id
string

Unique response identifier.

​
object
string
Example:

"chat.completion"

​
created
integer<int64>

Creation time as a Unix timestamp (seconds).

​
model
string

Model name used.

​
choices
object[]
​
usage
object