OpenAI 相容介面升級：深度支援 Claude

我們已針對 Claude 系列模型，對 OpenAI 相容介面進行更深入的最佳化。現在，你可以更精準、更方便地控制思考與快取。多輪對話中的交錯思考也變得更容易使用，無須額外參數即可無縫整合。此外，介面也支援啟用 Anthropic 提供的 Beta 功能。

1. 模型思考（延伸思考）

1.1 交錯思考的優勢

未啟用交錯思考時，模型只會在助理回合開始時思考一次；收到工具結果後，便會直接產生後續回覆，不會再產生新的思考區塊：

User → [Thinking] → Tool Call → Tool Result → Response

啟用交錯思考後，模型每次收到工具結果時都會插入新的思考區塊，形成推理鏈：

User → [Thinking] → Tool Call → Tool Result → [Thinking] → Response
                                                ↑ Interleaved Thinking

這讓模型能夠：

• 根據工具結果進行二次推理，而不是單純串接輸出。
• 在多次工具呼叫之間串接推理，每個決策都以先前步驟的分析為依據。

參考資料：Anthropic 交錯式思考

1.2 啟用思考

你可以透過以下四種方式啟用思考，任選一種即可：

優先順序（同時使用多種方式時）：reasoning_effort > reasoning.max_tokens > reasoning.effort > -think 後綴

可用的推理強度值：minimal / low / medium / high / xhigh

1.3 思考內容回傳

回應訊息會包含兩個新欄位：

• reasoning_content：思考內容（字串），方便顯示。
• reasoning_details：思考內容的完整結構化資訊；在多輪對話中必須原樣回傳，不同供應商的內部結構可能有所不同。

非串流範例（省略無關欄位）：

{
  "choices": [{
    "message": {
      "role": "assistant",
      "content": "Hello! How can I help you today?",
      "reasoning_content": "The user is just saying hello...",
      "reasoning_details": {
        "type": "thinking",
        "thinking": "The user is just saying hello...",
        "signature": "Er8CCkYI..."
      }
    }
  }]
}

串流回應中，思考內容會透過 delta.reasoning_content 和 delta.reasoning_details 分段傳送。完整的串流串接邏輯請參閱下方完整範例。

1.4 在多輪對話中保留思考內容（交錯思考為內建功能，無須額外參數）

若要讓模型在多輪對話中延續推理能力，只需將先前回傳的 reasoning_details原樣放入下一輪的 assistant 訊息中：

messages = [
    {"role": "user", "content": "What's the weather like in Boston?"},
    {
        "role": "assistant",
        "content": response.choices[0].message.content,
        "tool_calls": response.choices[0].message.tool_calls,
        "reasoning_details": response.choices[0].message.reasoning_details,
    },
    {
        "role": "tool",
        "tool_call_id": "toolu_xxx",
        "content": '{"temperature": 45, "condition": "rainy"}',
    }
]

AIHubMix 偵測到請求中包含歷史思考資訊時，會自動啟用交錯思考，讓模型收到工具呼叫結果後無須額外參數即可繼續深度推理。

1.5 完整範例

以下兩個範例展示完整的多輪 Tool Call + 交錯思考流程：使用者提問 → 模型思考並呼叫工具 → 注入工具結果（保留 reasoning_details）→ 模型透過交錯思考給出最終回覆。

非串流 · 交錯思考

import os
import json
from openai import OpenAI

client = OpenAI(
    base_url="https://aihubmix.com/v1",
    api_key=os.environ.get("AIHUBMIX_API_KEY", "sk-***"),
)

# ── Tool definition ───────────────────────────────────────────
tools = [{
    "type": "function",
    "function": {
        "name": "get_weather",
        "description": "Get current weather for a location",
        "parameters": {
            "type": "object",
            "properties": {"location": {"type": "string", "description": "City name"}},
            "required": ["location"]
        }
    }
}]

# ── Mock tool execution ───────────────────────────────────────
WEATHER_DB = {
    "boston": {"temperature": "45°F (7°C)", "condition": "rainy", "humidity": "85%", "wind": "15 mph NE"},
    "tokyo":  {"temperature": "72°F (22°C)", "condition": "sunny", "humidity": "45%", "wind": "5 mph S"},
}

def execute_tool(name: str, args: dict) -> str:
    if name == "get_weather":
        key = next((k for k in WEATHER_DB if k in args.get("location", "").lower()), None)
        return json.dumps(WEATHER_DB.get(key, {"temperature": "65°F", "condition": "clear"}))
    return "{}"

# ── Multi-turn conversation loop ─────────────────────────────
messages = [
    {"role": "user", "content": "What's the weather like in Boston? Then recommend what to wear."}
]

turn = 0
while True:
    turn += 1
    print(f"\n── Turn {turn} ──")

    response = client.chat.completions.create(
        model="claude-sonnet-4-5",
        messages=messages,
        tools=tools,
        extra_body={"reasoning": {"max_tokens": 2000}},
    )
    msg = response.choices[0].message

    # Print thinking process
    if msg.reasoning_content:
        label = "Interleaved Thinking" if turn > 1 else "Thinking"
        print(f"[{label}] {msg.reasoning_content}")

    # Print response content
    if msg.content:
        print(f"[Response] {msg.content}")

    # Print tool calls
    if msg.tool_calls:
        for tc in msg.tool_calls:
            print(f"[Tool Call: {tc.function.name}] {tc.function.arguments}")

    # Build assistant message, preserve reasoning_details (critical!)
    assistant_msg = {"role": "assistant", "content": msg.content}
    if msg.tool_calls:
        assistant_msg["tool_calls"] = msg.tool_calls
    if msg.reasoning_details:
        assistant_msg["reasoning_details"] = msg.reasoning_details  # pass back unmodified
    messages.append(assistant_msg)

    # No tool_calls means conversation is done
    if not msg.tool_calls:
        break

    # Execute tools and append results to messages
    for tc in msg.tool_calls:
        args = json.loads(tc.function.arguments)
        result = execute_tool(tc.function.name, args)
        print(f"[Tool Result: {tc.function.name}] {result}")
        messages.append({"role": "tool", "tool_call_id": tc.id, "content": result})

串流 · 交錯思考

import os
import sys
import json
from openai import OpenAI

client = OpenAI(
    base_url="https://aihubmix.com/v1",
    api_key=os.environ.get("AIHUBMIX_API_KEY", "sk-***"),
)

# ── Tool definition & mock execution ─────────────────────────
tools = [{
    "type": "function",
    "function": {
        "name": "get_weather",
        "description": "Get current weather for a location",
        "parameters": {
            "type": "object",
            "properties": {"location": {"type": "string", "description": "City name"}},
            "required": ["location"]
        }
    }
}]

WEATHER_DB = {
    "boston": {"temperature": "45°F (7°C)", "condition": "rainy", "humidity": "85%", "wind": "15 mph NE"},
    "tokyo":  {"temperature": "72°F (22°C)", "condition": "sunny", "humidity": "45%", "wind": "5 mph S"},
}

def execute_tool(name: str, args: dict) -> str:
    if name == "get_weather":
        key = next((k for k in WEATHER_DB if k in args.get("location", "").lower()), None)
        return json.dumps(WEATHER_DB.get(key, {"temperature": "65°F", "condition": "clear"}))
    return "{}"

# ── Stream response collector ────────────────────────────────
def stream_and_collect(turn: int, **kwargs):
    """Stream response, print thinking/content in real-time, accumulate reasoning_details/tool_calls."""
    rd = {}            # accumulated reasoning_details
    content = ""       # accumulated response text
    tc_map = {}        # accumulated tool_calls (by index)
    cur = "none"       # current output section: none / thinking / content

    stream = client.chat.completions.create(stream=True, **kwargs)
    for chunk in stream:
        if not chunk.choices:
            continue
        delta = chunk.choices[0].delta

        # ── Handle thinking ──
        rd_delta = getattr(delta, "reasoning_details", None)
        if rd_delta and isinstance(rd_delta, dict):
            for k, v in rd_delta.items():
                if k == "type":
                    rd[k] = v
                elif isinstance(v, str):
                    rd[k] = rd.get(k, "") + v
                elif v is not None:
                    rd[k] = v
            # Print thinking chunks in real-time
            thinking_chunk = rd_delta.get("thinking", "")
            if thinking_chunk:
                if cur != "thinking":
                    cur = "thinking"
                    label = "Interleaved Thinking" if turn > 1 else "Thinking"
                    sys.stdout.write(f"\n[{label}] ")
                sys.stdout.write(thinking_chunk)
                sys.stdout.flush()

        # ── Handle content ──
        if delta.content:
            if cur != "content":
                if cur == "thinking":
                    sys.stdout.write("\n")
                cur = "content"
                sys.stdout.write("\n[Response] ")
            sys.stdout.write(delta.content)
            sys.stdout.flush()
            content += delta.content

        # ── Handle tool_calls ──
        for tc in delta.tool_calls or []:
            i = tc.index
            if i not in tc_map:
                tc_map[i] = {"id": "", "type": "function",
                             "function": {"name": "", "arguments": ""}}
            if tc.id:
                tc_map[i]["id"] = tc.id
            if tc.function:
                tc_map[i]["function"]["name"] += tc.function.name or ""
                tc_map[i]["function"]["arguments"] += tc.function.arguments or ""

    # End current output section
    if cur in ("thinking", "content"):
        sys.stdout.write("\n")

    tool_calls = [tc_map[i] for i in sorted(tc_map)] if tc_map else None
    return {
        "content": content or None,
        "reasoning_details": rd or None,
        "tool_calls": tool_calls,
    }

# ── Multi-turn conversation loop ─────────────────────────────
messages = [
    {"role": "user", "content": "What's the weather like in Boston? Then recommend what to wear."}
]

turn = 0
while True:
    turn += 1
    print(f"\n── Turn {turn} ──")

    result = stream_and_collect(
        turn,
        model="claude-sonnet-4-5",
        messages=messages,
        tools=tools,
        extra_body={"reasoning": {"max_tokens": 2000}},
    )

    # Print tool calls
    if result["tool_calls"]:
        for tc in result["tool_calls"]:
            print(f"[Tool Call: {tc['function']['name']}] {tc['function']['arguments']}")

    # Build assistant message, preserve reasoning_details (critical!)
    assistant_msg = {"role": "assistant", "content": result["content"]}
    if result["tool_calls"]:
        assistant_msg["tool_calls"] = result["tool_calls"]
    if result["reasoning_details"]:
        assistant_msg["reasoning_details"] = result["reasoning_details"]  # pass back unmodified
    messages.append(assistant_msg)

    # No tool_calls means conversation is done
    if not result["tool_calls"]:
        break

    # Execute tools and append results to messages
    for tc in result["tool_calls"]:
        args = json.loads(tc["function"]["arguments"])
        tool_result = execute_tool(tc["function"]["name"], args)
        print(f"[Tool Result: {tc['function']['name']}] {tool_result}")
        messages.append({"role": "tool", "tool_call_id": tc["id"], "content": tool_result})

1.6 思考強度對應規則

推理強度模式：

• Opus 4.6 / Sonnet 4.6 及以上版本：對應至 Anthropic 原生的 Adaptive Thinking 推理強度。
• 其他模型：使用 budget_tokens 公式計算：

budget_tokens = max(min(max_tokens × effort_ratio, 128000), 1024)

Adaptive Thinking 推理強度對應：

max_tokens 模式：直接指定為 Anthropic 的 budget_tokens。

-think 後綴：Opus/Sonnet 4.6+ 使用 adaptive thinking（effort=medium）；其他模型則設定 budget_tokens = min(10240, max_tokens - 1)，max_tokens 預設為 4096。

2. 提示詞快取

透過 Chat 介面向 Claude 模型發出請求時，可以使用提示詞快取。在訊息中設定 cache_control 快取斷點，即可快取大型文字區塊（例如角色卡、RAG 資料、書籍章節等）以供重複使用，讓後續請求直接命中快取，大幅降低成本。

Claude 官方文件：提示詞快取

2.1 快取費用

2.2 支援的模型與快取長度下限

斷點數量限制：每個請求最多可設定 4 個 cache_control 斷點。

2.3 快取 TTL

1 小時 TTL 的寫入成本較高，但在長時間工作階段中，減少重複寫入即可節省整體費用。所有供應商（包括 Anthropic、Amazon Bedrock、Google Vertex AI）提供的 Claude 4.5 及後續版本模型，皆支援 1 小時 TTL。

2.4 使用方式

你可以在 system、user（包括圖片）和 tools 中使用 cache_control 欄位設定快取斷點。以下範例僅展示關鍵結構，省略大型文字區塊。

System 訊息快取（預設 5 分鐘 TTL）：

{
  "model": "claude-opus-4-5",
  "messages": [
    {
      "role": "system",
      "content": [
        {"type": "text", "text": "You are an AI assistant"},
        {
          "type": "text",
          "text": "(long context)",
          "cache_control": {"type": "ephemeral"}
        }
      ]
    },
    {
      "role": "user",
      "content": [{"type": "text", "text": "Hello"}]
    }
  ]
}

User 訊息快取（1 小時 TTL）：

{
  "model": "claude-opus-4-5",
  "messages": [
    {
      "role": "system",
      "content": [{"type": "text", "text": "You are an AI assistant"}]
    },
    {
      "role": "user",
      "content": [
        {
          "type": "text",
          "text": "(long context)",
          "cache_control": {"type": "ephemeral", "ttl": "1h"}
        },
        {"type": "text", "text": "Hello"}
      ]
    }
  ]
}

圖片訊息快取：

{
  "role": "user",
  "content": [
    {
      "type": "image_url",
      "image_url": {"detail": "auto", "url": "data:image/jpeg;base64,/9j/4AAQ..."},
      "cache_control": {"type": "ephemeral"}
    },
    {"type": "text", "text": "What's this?"}
  ]
}

工具定義快取：

cache_control 放在工具物件的頂層（與 type 和 function 同一層）：

{
  "tools": [{
    "type": "function",
    "function": {
      "name": "get_weather",
      "description": "Get current weather for a location",
      "parameters": {
        "type": "object",
        "properties": {"city": {"type": "string"}},
        "required": ["city"]
      }
    },
    "cache_control": {"type": "ephemeral", "ttl": "1h"}
  }]
}

2.5 查看快取狀態

回應中的 usage 會回傳 claude_cache_tokens_details，記錄快取的詳細資訊：

首次請求（建立快取）：

{
  "usage": {
    "prompt_tokens": 22,
    "completion_tokens": 890,
    "total_tokens": 912,
    "claude_cache_tokens_details": {
      "cache_creation_input_tokens": 6266,
      "cache_read_input_tokens": 0,
      "cache_write_5_minutes_input_tokens": 6266,
      "cache_write_1_hour_input_tokens": 0
    }
  }
}

後續請求（快取命中）：

{
  "usage": {
    "prompt_tokens": 22,
    "completion_tokens": 810,
    "total_tokens": 832,
    "prompt_tokens_details": {
      "cached_tokens": 6266
    },
    "claude_cache_tokens_details": {
      "cache_creation_input_tokens": 0,
      "cache_read_input_tokens": 6266,
      "cache_write_5_minutes_input_tokens": 0,
      "cache_write_1_hour_input_tokens": 0
    }
  }
}

3. anthropic-beta 請求標頭

你可以透過 HTTP 標頭 anthropic-beta 啟用 Claude 模型的 Beta 功能，AIHubMix 會將此標頭轉送至 Anthropic API。

使用方式

在請求標頭中加入 anthropic-beta，並將值設為對應的 Beta 功能識別碼：

curl "https://aihubmix.com/v1/chat/completions" \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AIHUBMIX_API_KEY" \
  -H "anthropic-beta: context-1m-2025-08-07" \
  -d '{
  "model": "claude-opus-4-5",
  "messages": [
    {
      "role": "system",
      "content": [
        {"type": "text", "text": "You are an AI assistant"},
        {
          "type": "text",
          "text": "(long context)",
          "cache_control": {"type": "ephemeral"}
        }
      ]
    },
    {"role": "user", "content": [{"type": "text", "text": "hello"}]}
  ]
}'

如需查詢可用的 Beta 識別碼，請參閱Anthropic API 文件。

最後更新：2026-06-01