Skip to main content
POST
/
v1
/
messages
cURL
curl --request POST \
  --url https://proxy.innk.cc/v1/messages \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "model": "claude-sonnet-4-5-20250929",
  "messages": [
    {
      "role": "user",
      "content": "Hello, how are you?"
    }
  ],
  "max_tokens": 4096,
  "stream": false,
  "temperature": 1,
  "top_k": 5,
  "top_p": 0.7
}
'
{
  "content": [
    {
      "text": "<string>",
      "type": "<string>"
    }
  ],
  "id": "msg_013Zva2CMHLNnXjNJJKqJ2EF",
  "model": "<string>",
  "role": "<string>",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "type": "message",
  "usage": {
    "input_tokens": 123,
    "output_tokens": 123
  }
}
The current document only lists the basic functions of Anthropic Messages, such as the Claude model. More parameter functions can be used, please refer toAnthropic API Reference。

Authorizations

​
Authorization
string
header
required

The authentication header format is Bearer <API_KEY>, where <API_KEY> is your API token.

Headers

​
anthropic-version
string
default:2023-06-01

Anthropic API version, only valid when using Claude model

​
anthropic-beta
string[]

Anthropic API beta, only valid when using Claude model

Body

application/json
​
model
string
required
Example:

"claude-sonnet-4-5-20250929"

​
messages
object[]
required
​
max_tokens
integer
default:4096
required

Maximum number of tokens

​
stream
boolean
default:false

Whether to stream conversations

​
temperature
number
default:1

temperature

​
top_k
integer

Top-K sampling

Example:

5

​
top_p
number

nuclear sampling

Example:

0.7

Response

200 - application/json

Model response

​
content
object[]
required
​
id
string
required
Example:

"msg_013Zva2CMHLNnXjNJJKqJ2EF"

​
model
string
required
​
role
string
required
​
stop_reason
enum<string>
required

Reason for stopping operation. This may be one of the following values:

  • "end_turn": The model reaches a natural stopping point
  • "max_tokens": We exceeded the requested max_tokens value or the model's maximum value
  • "stop_sequence": Generates one of the custom stop_sequences you provide
  • "tool_use": The model calls one or more tools
  • "pause_turn": We paused a long running turn. You can provide this response directly in subsequent requests to allow the model to continue running.
  • "refusal": when the streaming classifier steps in to handle a potential policy violation

In non-streaming mode, this value is never empty. In streaming mode, null in the message_start event, non-null otherwise.

Available options:
end_turn,
max_tokens,
stop_sequence,
tool_use,
pause_turn,
refusal
​
stop_sequence
null
required

Did you generate a custom stop sequence?

If so, this value is a non-empty string. If a custom stop sequence was generated, this value is a non-empty string.

Example:

null

​
type
string
required
Example:

"message"

​
usage
object
required

Dosage