自適應思考

自適應思考是在 Claude Opus 4.6 上使用延伸思考的推薦方式。自適應思考不需要手動設定思考 token 預算，而是讓 Claude 根據每個請求的複雜度動態決定何時以及思考多少。

支援的模型

自適應思考支援以下模型：

• Claude Opus 4.6 (claude-opus-4-6)

自適應思考的運作方式

在自適應模式中，思考對模型來說是可選的。Claude 會評估每個請求的複雜度，並決定是否以及思考多少。在預設的 effort 等級（high）下，Claude 幾乎總是會思考。在較低的 effort 等級下，Claude 可能會跳過較簡單問題的思考。

自適應思考也會自動啟用交錯思考。這意味著 Claude 可以在工具呼叫之間進行思考，使其在代理工作流程中特別有效。

如何使用自適應思考

在您的 API 請求中將 thinking.type 設定為 "adaptive"：

自適應思考搭配 effort 參數

您可以將自適應思考與 effort 參數結合使用，以引導 Claude 進行多少思考。effort 等級作為 Claude 思考分配的軟性指引：

自適應思考的串流

自適應思考與串流無縫配合。思考區塊透過 thinking_delta 事件進行串流，就像手動思考模式一樣：

自適應 vs 手動 vs 停用思考

重要注意事項

驗證變更

使用自適應思考時，先前的助理回合不需要以思考區塊開頭。這比手動模式更靈活，手動模式中 API 會強制要求啟用思考的回合以思考區塊開頭。

提示快取

使用 adaptive 思考的連續請求會保留提示快取斷點。然而，在 adaptive 和 enabled/disabled 思考模式之間切換會破壞訊息的快取斷點。無論模式如何變更，系統提示和工具定義都會保持快取。

調整思考行為

自適應思考的觸發行為是可透過提示引導的。如果 Claude 思考的頻率比您期望的多或少，您可以在系統提示中添加指引：

Extended thinking adds latency and should only be used when it
will meaningfully improve answer quality — typically for problems
that require multi-step reasoning. When in doubt, respond directly.

成本控制

使用 max_tokens 作為總輸出（思考 + 回應文字）的硬性限制。effort 參數提供額外的軟性指引，控制 Claude 分配多少思考。兩者結合使用，可以有效控制成本。

在 high 和 max effort 等級下，Claude 可能會進行更廣泛的思考，並且更可能耗盡 max_tokens 預算。如果您在回應中觀察到 stop_reason: "max_tokens"，請考慮增加 max_tokens 以給模型更多空間，或降低 effort 等級。

處理思考區塊

以下概念適用於所有支援延伸思考的模型，無論您使用自適應還是手動模式。

摘要思考

With extended thinking enabled, the Messages API for Claude 4 models returns a summary of Claude's full thinking process. Summarized thinking provides the full intelligence benefits of extended thinking, while preventing misuse.

Here are some important considerations for summarized thinking:

• You're charged for the full thinking tokens generated by the original request, not the summary tokens.
• The billed output token count will not match the count of tokens you see in the response.
• The first few lines of thinking output are more verbose, providing detailed reasoning that's particularly helpful for prompt engineering purposes.
• As Anthropic seeks to improve the extended thinking feature, summarization behavior is subject to change.
• Summarization preserves the key ideas of Claude's thinking process with minimal added latency, enabling a streamable user experience and easy migration from Claude Sonnet 3.7 to Claude 4 and later models.
• Summarization is processed by a different model than the one you target in your requests. The thinking model does not see the summarized output.

思考加密

Full thinking content is encrypted and returned in the signature field. This field is used to verify that thinking blocks were generated by Claude when passed back to the API.

Here are some important considerations on thinking encryption:

• When streaming responses, the signature is added via a signature_delta inside a content_block_delta event just before the content_block_stop event.
• signature values are significantly longer in Claude 4 models than in previous models.
• The signature field is an opaque field and should not be interpreted or parsed - it exists solely for verification purposes.
• signature values are compatible across platforms (Claude APIs, Amazon Bedrock, and Vertex AI). Values generated on one platform will be compatible with another.

思考編輯

Occasionally Claude's internal reasoning will be flagged by our safety systems. When this occurs, we encrypt some or all of the thinking block and return it to you as a redacted_thinking block. redacted_thinking blocks are decrypted when passed back to the API, allowing Claude to continue its response without losing context.

When building customer-facing applications that use extended thinking:

• Be aware that redacted thinking blocks contain encrypted content that isn't human-readable
• Consider providing a simple explanation like: "Some of Claude's internal reasoning has been automatically encrypted for safety reasons. This doesn't affect the quality of responses."
• If showing thinking blocks to users, you can filter out redacted blocks while preserving normal thinking blocks
• Be transparent that using extended thinking features may occasionally result in some reasoning being encrypted
• Implement appropriate error handling to gracefully manage redacted thinking without breaking your UI

Here's an example showing both normal and redacted thinking blocks:

{
  "content": [
    {
      "type": "thinking",
      "thinking": "Let me analyze this step by step...",
      "signature": "WaUjzkypQ2mUEVM36O2TxuC06KN8xyfbJwyem2dw3URve/op91XWHOEBLLqIOMfFG/UvLEczmEsUjavL...."
    },
    {
      "type": "redacted_thinking",
      "data": "EmwKAhgBEgy3va3pzix/LafPsn4aDFIT2Xlxh0L5L8rLVyIwxtE3rAFBa8cr3qpPkNRj2YfWXGmKDxH4mPnZ5sQ7vB9URj2pLmN3kF8/dW5hR7xJ0aP1oLs9yTcMnKVf2wRpEGjH9XZaBt4UvDcPrQ..."
    },
    {
      "type": "text",
      "text": "Based on my analysis..."
    }
  ]
}

When passing thinking and redacted_thinking blocks back to the API in a multi-turn conversation, you must include the complete unmodified block back to the API for the last assistant turn. This is critical for maintaining the model's reasoning flow. We suggest always passing back all thinking blocks to the API. For more details, see the Preserving thinking blocks section.

定價

For complete pricing information including base rates, cache writes, cache hits, and output tokens, see the pricing page.

The thinking process incurs charges for:

• Tokens used during thinking (output tokens)
• Thinking blocks from the last assistant turn included in subsequent requests (input tokens)
• Standard text output tokens

When using summarized thinking:

• Input tokens: Tokens in your original request (excludes thinking tokens from previous turns)
• Output tokens (billed): The original thinking tokens that Claude generated internally
• Output tokens (visible): The summarized thinking tokens you see in the response
• No charge: Tokens used to generate the summary

其他主題

延伸思考頁面以更詳細的方式涵蓋了幾個主題，並附有特定模式的程式碼範例：

• 搭配思考的工具使用：自適應思考適用相同的規則 — 在工具呼叫之間保留思考區塊，並注意思考啟用時 tool_choice 的限制。
• 提示快取：使用自適應思考時，使用相同思考模式的連續請求會保留快取斷點。在 adaptive 和 enabled/disabled 模式之間切換會破壞訊息的快取斷點（系統提示和工具定義會保持快取）。
• 上下文視窗：思考 token 如何與 max_tokens 和上下文視窗限制互動。

後續步驟