Chat (prompt cache)

POST

/v1/messages

Prompt cache mode can be supported by setting parameters: "cache_control": {"type": "ephemeral"}
(If you need a dedicated channel, please contact the administrator)

Request

Authorization

Provide your bearer token in the

Authorization

header when making requests to protected resources.

Example:

Authorization: Bearer ********************

Header Params

Body Params application/json

Example

{
  "model": "claude-3-7-sonnet-20250219",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Summarize the content of the article, no more than 50 words: {text}",
          "cache_control": {
            "type": "ephemeral"
          }
        }
      ]
    }
  ]
}

Request Code Samples

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location '/v1/messages' \
--header 'x-api-key: {YOUR_API_KEY}' \
--header 'Authorization: Bearer <token>' \
--header 'content-type: application/json' \
--data '{
  "model": "claude-3-7-sonnet-20250219",
  "messages": [
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "Summarize the content of the article, no more than 50 words: {text}",
          "cache_control": {
            "type": "ephemeral"
          }
        }
      ]
    }
  ]
}'

Responses

🟢200OK

application/json

Bodyapplication/json

Example

{
  "id": "msg_bdrk_01MKgCTVgyqRZTHF9ej4xwZ8",
  "type": "message",
  "role": "assistant",
  "model": "claude-3-7-sonnet-20250219",
  "content": [
    {
      "type": "text",
      "text": "The paper points out that AI hallucinations are an inevitable product of the training mechanism, and the model tends to make blind guesses rather than admit not knowing. Comparative tests show that the accuracy of the model that is good at guessing seems to be high, but the error rate is extremely high; while the honest model would rather not answer than tell nonsense."
    }
  ],
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 3,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 1869,
    "output_tokens": 93
  }
}

Modified at 2026-07-09 10:12:26

Chat

Streaming Response