# GPT 结构化输出与 PDF 输入：JSON Schema、response_format

Source: https://ofox.ai/zh/docs/develop/models/gpt/structured-output

本页说明 GPT 系列的结构化输出与 PDF 输入写法；模型 ID 与接口配置见 [GPT 系列概览](https://ofox.ai/zh/docs/develop/models/gpt)。

## 结构化输出
通过 `response_format` 传入 JSON Schema，`strict: true` 时输出严格符合 Schema：

```python filename="gpt_json_schema.py"
import json
import os
from openai import OpenAI

client = OpenAI(base_url="https://api.ofox.ai/v1", api_key=os.environ["OFOX_API_KEY"])

resp = client.chat.completions.create(
    model="openai/gpt-5.4-mini",
    response_format={
        "type": "json_schema",
        "json_schema": {
            "name": "price",
            "strict": True,
            "schema": {
                "type": "object",
                "properties": {"price": {"type": "number"}, "currency": {"type": "string"}},
                "required": ["price", "currency"],
                "additionalProperties": False,
            },
        },
    },
    messages=[{"role": "user", "content": "球拍和球共 1.10 美元，球拍比球贵 1 美元，球多少钱？"}],
)
print(json.loads(resp.choices[0].message.content))
```

更多写法见[结构化输出](https://ofox.ai/zh/docs/develop/guides/structured-output)。

## PDF 输入

**Chat Completions**

```python filename="gpt_pdf.py"
import base64
import os
from openai import OpenAI

client = OpenAI(base_url="https://api.ofox.ai/v1", api_key=os.environ["OFOX_API_KEY"])
pdf_b64 = base64.b64encode(open("invoice.pdf", "rb").read()).decode()

resp = client.chat.completions.create(
    model="openai/gpt-5.4-mini",
    messages=[{"role": "user", "content": [
        {"type": "file", "file": {"filename": "invoice.pdf", "file_data": f"data:application/pdf;base64,{pdf_b64}"}},
        {"type": "text", "text": "这张发票的总金额是多少？"},
    ]}],
)
print(resp.choices[0].message.content)
```

**Responses**

```python filename="gpt_pdf_responses.py"
import base64
import os
from openai import OpenAI

client = OpenAI(base_url="https://api.ofox.ai/v1", api_key=os.environ["OFOX_API_KEY"])
pdf_b64 = base64.b64encode(open("invoice.pdf", "rb").read()).decode()

resp = client.responses.create(
    model="openai/gpt-5.4-mini",
    input=[{"role": "user", "content": [
        {"type": "input_file", "filename": "invoice.pdf", "file_data": f"data:application/pdf;base64,{pdf_b64}"},
        {"type": "input_text", "text": "这张发票的总金额是多少？"},
    ]}],
)
print(resp.output_text)
```

其他报错见[常见报错](https://ofox.ai/zh/docs/develop/models/gpt#errors)。