> ## Documentation Index
> Fetch the complete documentation index at: https://docs.eztokens.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# 推理思考

> 扩展思维和逐步推理 Token

部分模型在生成最终响应之前会产生中间推理（或「思考」）Token。支持推理的模型包括 Anthropic 的 Claude（扩展思考）和 DeepSeek-R1。

## 计费

思考 Token 按输出 Token 同价计费。推理无单独定价。

## 设置推理强度

<Tabs>
  <Tab title="cURL">
    ```bash theme={null}
    curl https://api.eztokens.ai/v1/chat/completions \
      -H "Content-Type: application/json" \
      -H "Authorization: Bearer $EZTOKENS_API_KEY" \
      -d '{
        "model": "deepseek-r1",
        "messages": [
          {"role": "user", "content": "Are there an infinite number of primes where n mod 4 == 3?"}
        ],
        "reasoning": {"effort": "high"}
      }'
    ```
  </Tab>

  <Tab title="Python">
    ```python theme={null}
    from openai import OpenAI

    client = OpenAI(
        api_key="EZTOKENS_API_KEY",
        base_url="https://api.eztokens.ai/v1"
    )

    response = client.chat.completions.create(
        model="deepseek-r1",
        messages=[
            {"role": "user", "content": "Are there an infinite number of primes where n mod 4 == 3?"}
        ],
        reasoning={"effort": "high"}
    )

    print(response.choices[0].message.content)
    ```
  </Tab>

  <Tab title="Node.js">
    ```javascript theme={null}
    import OpenAI from "openai";

    const client = new OpenAI({
      apiKey: "EZTOKENS_API_KEY",
      baseURL: "https://api.eztokens.ai/v1",
    });

    const response = await client.chat.completions.create({
      model: "deepseek-r1",
      messages: [
        { role: "user", content: "Are there an infinite number of primes where n mod 4 == 3?" }
      ],
      reasoning: { effort: "high" },
    });

    console.log(response.choices[0].message.content);
    ```
  </Tab>
</Tabs>

## 响应示例

## 流式推理

推理同样支持流式输出 — 设置 `stream: true` 即可。推理内容在最终 `delta.content` 之前以 `delta.reasoning_content` 的形式按 chunk 抵达。

```python theme={null}
from openai import OpenAI

client = OpenAI(
    api_key="EZTOKENS_API_KEY",
    base_url="https://api.eztokens.ai/v1"
)

stream = client.chat.completions.create(
    model="deepseek-r1",
    messages=[{"role": "user", "content": "Solve: x^2 - 5x + 6 = 0"}],
    reasoning={"effort": "high"},
    stream=True
)

for chunk in stream:
    if chunk.choices[0].delta.content:
        print(chunk.choices[0].delta.content, end="")
```