建立訊息

POST

messages

import os
import anthropic

client = anthropic.Anthropic(
    base_url="https://api.cometapi.com",
    api_key=os.environ["COMETAPI_KEY"],
)

message = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=1024,
    system="You are a helpful assistant.",
    messages=[
        {"role": "user", "content": "Hello, world"}
    ],
)

print(message.content[0].text)

{
  "id": "<string>",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "text": "<string>",
      "thinking": "<string>",
      "signature": "<string>",
      "id": "<string>",
      "name": "<string>",
      "input": {}
    }
  ],
  "model": "<string>",
  "stop_sequence": "<string>",
  "usage": {
    "input_tokens": 123,
    "output_tokens": 123,
    "cache_creation_input_tokens": 123,
    "cache_read_input_tokens": 123,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 123,
      "ephemeral_1h_input_tokens": 123
    },
    "output_tokens_details": {
      "thinking_tokens": 123
    }
  }
}

CometAPI 原生支援 Anthropic 訊息 API，讓你能直接存取 Claude 模型及其 Anthropic 專屬功能。對於自適應思考、Prompt 快取與 effort 控制等 Claude 能力，請使用此端點。

請以官方的 Anthropic Messages API reference 作為完整參數清單、回應 schema 與 Claude 專屬行為的權威來源。此 CometAPI 頁面說明如何透過 CometAPI 傳送該請求格式。

隨著 Claude 功能持續演進，Anthropic 的請求參數與回應欄位可能會變動。請查看 Anthropic 訊息 API 文件以取得最新且完整的參數清單與供應商特定行為。

許多較新的 Claude 模型在 Messages API 上會拒絕非預設的 temperature、top_p 與 top_k 值。除非你已確認所選 model 支援，否則請省略這些取樣欄位。若 model 回傳不支援或已棄用參數的錯誤，請從請求中移除該欄位。

驗證同時支援 x-api-key 與 Authorization: Bearer 標頭。官方 Anthropic SDK 預設使用 x-api-key。

快速開始

若要搭配 CometAPI 使用官方 Anthropic SDK，請設定 base URL：

import os
import anthropic

client = anthropic.Anthropic(
    base_url="https://api.cometapi.com",
    api_key=os.environ["COMETAPI_KEY"],
)

message = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=1024,
    messages=[{"role": "user", "content": "Hello!"}],
)
print(message.content[0].text)

import Anthropic from "@anthropic-ai/sdk";

const client = new Anthropic({
    apiKey: process.env.COMETAPI_KEY,
    baseURL: "https://api.cometapi.com",
});

const message = await client.messages.create({
    model: "claude-sonnet-5",
    max_tokens: 1024,
    messages: [{ role: "user", content: "Hello!" }],
});
console.log(message.content[0].text);

控制自適應思考

搭配 output_config.effort 使用自適應思考，以控制 Claude 在回應上投入多少工作。較新的 Claude 模型會拒絕舊版的手動 thinking 結構 thinking={"type": "enabled", "budget_tokens": ...}。

message = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=4096,
    thinking={"type": "adaptive"},
    output_config={"effort": "xhigh"},
    messages=[
        {
            "role": "user",
            "content": "Analyze the trade-offs between a monolithic architecture and microservices for a small engineering team.",
        }
    ],
)

for block in message.content:
    if block.type == "text":
        print(block.text)

使用較高的 effort 等級時，thinking tokens 會計入你的 max_tokens 限制。請將 max_tokens 設得足夠高，以同時容納 thinking 與最終答案。

快取 Prompt

為了降低後續請求的延遲與成本，可將大型 system prompts 或對話前綴進行快取。將 cache_control 加到應快取的內容區塊：

message = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=1024,
    system=[
        {
            "type": "text",
            "text": "You are an expert code reviewer. [Long detailed instructions...]",
            "cache_control": {"type": "ephemeral"},
        }
    ],
    messages=[{"role": "user", "content": "Review this code..."}],
)

快取使用情況會在回應的 usage 欄位中回報：

cache_creation_input_tokens — 寫入快取的 tokens（以較高費率計費）
cache_read_input_tokens — 從快取讀取的 tokens（以較低費率計費）

Prompt 快取要求被快取的內容區塊至少有 1,024 tokens。短於此長度的內容將不會被快取。

串流回應

若要使用 Server-Sent Events (SSE) 進行串流回應，請設定 stream: true。事件會依照以下順序到達：

message_start — 包含訊息中繼資料與初始 usage
content_block_start — 標記每個內容區塊的開始
content_block_delta — 遞增的文字片段（text_delta）
content_block_stop — 標記每個內容區塊的結束
message_delta — 最終的 stop_reason 與完整的 usage
message_stop — 表示串流結束

with client.messages.stream(
    model="claude-sonnet-5",
    max_tokens=256,
    messages=[{"role": "user", "content": "Hello"}],
) as stream:
    for text in stream.text_stream:
        print(text, end="")

控制 effort

若要控制 Claude 在產生回應時投入多少 effort，請使用 output_config.effort：

message = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=4096,
    messages=[
        {"role": "user", "content": "Summarize this briefly."}
    ],
    output_config={"effort": "low"},  # "low", "medium", "high", "xhigh", or "max"
)

使用伺服器工具

Claude 支援在 Anthropic 基礎設施上執行的伺服器端工具：

Web Fetch
Web Search

從 URL 擷取並分析內容：

message = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "Analyze the content at https://arxiv.org/abs/1512.03385"}
    ],
    tools=[
        {"type": "web_fetch_20250910", "name": "web_fetch", "max_uses": 5}
    ],
)

搜尋網路上的即時資訊：

message = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=1024,
    messages=[
        {"role": "user", "content": "What are the latest developments in AI?"}
    ],
    tools=[
        {"type": "web_search_20250305", "name": "web_search", "max_uses": 5}
    ],
)

回應範例

來自 CometAPI Anthropic 端點的典型回應：

{
  "id": "msg_bdrk_01UjHdmSztrL7QYYm7CKBDFB",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "type": "text",
      "text": "Hello!"
    }
  ],
  "model": "claude-sonnet-5",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "usage": {
    "input_tokens": 19,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 0
    },
    "output_tokens": 4
  }
}

與 OpenAI 相容端點比較

功能	Anthropic 訊息 (`/v1/messages`)	OpenAI 相容 (`/v1/chat/completions`)
自適應思考	`thinking` 搭配 `type: "adaptive"` 與 `output_config.effort`	不提供
Prompt 快取	內容區塊上的 `cache_control`	不提供
努力程度控制	`output_config.effort`	不提供
Web 擷取/搜尋	伺服器工具（`web_fetch`、`web_search`）	不提供
驗證標頭	`x-api-key` 或 `Bearer`	僅 `Bearer`
回應格式	Anthropic 格式（`content` 區塊）	OpenAI 格式（`choices`、`message`）
模型	僅 Claude	多供應商（GPT、Claude、Gemini 等）

授權

x-api-key

string

header

必填

Your CometAPI key passed via the x-api-key header. Authorization: Bearer $COMETAPI_KEY is also supported.

標頭

anthropic-version

string

預設值:2023-06-01

The Anthropic API version to use. Defaults to 2023-06-01.

範例:

"2023-06-01"

anthropic-beta

string

Comma-separated list of beta features to enable. Examples: max-tokens-3-5-sonnet-2024-07-15, pdfs-2024-09-25, output-128k-2025-02-19.

主體

application/json

model

string

必填

The Claude model to use. See the Models page for available Claude model IDs.

範例:

"claude-sonnet-5"

messages

object[]

必填

The conversation messages. Must alternate between user and assistant roles. Each message's content can be a string or an array of content blocks (text, image, document, tool_use, tool_result). There is a limit of 100,000 messages per request.

Show child attributes

max_tokens

integer

必填

The maximum number of tokens to generate. The model may stop before reaching this limit. When using thinking, the thinking tokens count towards this limit.

必填範圍: x >= 1

範例:

1024

system

System prompt providing context and instructions to Claude. Can be a plain string or an array of content blocks (useful for prompt caching).

temperature

number

預設值:1

Model-dependent sampling control. Many newer Claude models reject non-default temperature values on the Messages API. Omit this field unless you have verified that the selected model accepts it; if the model returns an unsupported or deprecated-parameter error, remove the field instead of substituting another sampling value.

必填範圍: 0 <= x <= 1

範例:

1

top_p

number

Model-dependent nucleus sampling control. Many newer Claude models reject non-default top_p values on the Messages API. Omit this field unless you have verified support for the selected model. Do not set temperature and top_p together.

必填範圍: 0 <= x <= 1

範例:

1

top_k

integer

Model-dependent top-k sampling control. Many newer Claude models reject non-default top_k values on the Messages API. Omit this field unless you have verified support for the selected model.

必填範圍: x >= 0

範例:

0

stream

boolean

預設值:false

If true, stream the response incrementally using Server-Sent Events (SSE). Events include message_start, content_block_start, content_block_delta, content_block_stop, message_delta, and message_stop.

stop_sequences

string[]

Custom strings that cause the model to stop generating when encountered. The stop sequence is not included in the response.

thinking

object

Controls Claude thinking when the selected model supports a configurable thinking mode. For newer adaptive-thinking models, use {"type":"adaptive"} with output_config.effort, or omit thinking when adaptive thinking is already the model default. Manual {"type":"enabled","budget_tokens":...} is supported only by older models and is rejected by newer Claude models.

Show child attributes

tools

object[]

Tools the model may use. Supports client-defined functions, web search (web_search_20250305), web fetch (web_fetch_20250910), code execution (code_execution_20250522), and more.

Show child attributes

tool_choice

object

Controls how the model uses tools.

Show child attributes

metadata

object

Request metadata for tracking and analytics.

Show child attributes

output_config

object

Configuration for response effort and output format. Field support depends on the selected Claude model.

Show child attributes

service_tier

enum<string>

The service tier to use. auto tries priority capacity first, standard_only uses only standard capacity.

可用選項:

auto,

standard_only

回應

200 - application/json

Successful response. When stream is true, the response is a stream of SSE events.

string

Unique identifier for this message (e.g., msg_01XFDUDYJgAACzvnptvVoYEL).

type

enum<string>

Always message.

可用選項:

message

role

enum<string>

Always assistant.

可用選項:

assistant

content

object[]

The response content blocks. May include text, thinking, tool_use, and other block types.

Show child attributes

model

string

The specific model version that generated this response, such as claude-sonnet-5.

stop_reason

enum<string>

Why the model stopped generating. refusal can be returned as a successful HTTP response when the model declines a request.

可用選項:

end_turn,

max_tokens,

stop_sequence,

tool_use,

pause_turn,

refusal

stop_sequence

string | null

The stop sequence that caused the model to stop, if applicable.

usage

object

Token usage statistics.

Show child attributes

建立模型回應

產生內容

import os
import anthropic

client = anthropic.Anthropic(
    base_url="https://api.cometapi.com",
    api_key=os.environ["COMETAPI_KEY"],
)

message = client.messages.create(
    model="claude-sonnet-5",
    max_tokens=1024,
    system="You are a helpful assistant.",
    messages=[
        {"role": "user", "content": "Hello, world"}
    ],
)

print(message.content[0].text)

{
  "id": "<string>",
  "type": "message",
  "role": "assistant",
  "content": [
    {
      "text": "<string>",
      "thinking": "<string>",
      "signature": "<string>",
      "id": "<string>",
      "name": "<string>",
      "input": {}
    }
  ],
  "model": "<string>",
  "stop_sequence": "<string>",
  "usage": {
    "input_tokens": 123,
    "output_tokens": 123,
    "cache_creation_input_tokens": 123,
    "cache_read_input_tokens": 123,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 123,
      "ephemeral_1h_input_tokens": 123
    },
    "output_tokens_details": {
      "thinking_tokens": 123
    }
  }
}

內容審核

API 金鑰

快速開始

控制自適應思考

快取 Prompt

串流回應

控制 effort

使用伺服器工具

回應範例

與 OpenAI 相容端點比較

授權

標頭

主體

回應

​快速開始

​控制自適應思考

​快取 Prompt

​串流回應

​控制 effort

​使用伺服器工具

​回應範例

​與 OpenAI 相容端點比較

授權

標頭

主體

回應

快速開始

控制自適應思考

快取 Prompt

串流回應

控制 effort

使用伺服器工具

回應範例

與 OpenAI 相容端點比較