FIM 补全

使用场景

在 FIM (Fill In the Middle) 补全任务中，用户输入希望保留的前缀和后缀内容，模型将根据提示生成中间缺失的部分。这种补全方式常见于代码自动补全、文本中间生成等应用场景。

使用方式

在 chat/completions 接口中使用，只需支持 model_id 设置为模型广场上的模型 id 即可，在模块卡片的右上角复制。

{
    "model": "model_id",
    "messages": "prompt",
    "params": "params",
    "extra_body": {"prefix": "前缀内容", "suffix": "可选的后缀内容"}
}

在 completions 接口中使用

{
    "model": "model_info",
    "prompt": "前缀内容",
    "suffix": "后缀内容"
}

使用示例

基于 OpenAI 的 chat.completions 接口使用FIM补全：

Python

from openai import OpenAI

client = OpenAI(
    api_key="NUWA_API_KEY", # 换成你在后台生成的 Key "sk-***"
    base_url="https://api.nuwaapi.com/v1"
)

messages = [
    {"role": "user", "content": "Please write a sum function code"},
]

response = client.chat.completions.create(
    model="gpt-4o-mini",
    messages=messages,
    extra_body={
            "prefix": f"""
def sum_numbers(numbers):
    # 如果列表为空，返回 0
    if not numbers:
        return 0
""",
            "suffix": f"""
# 测试
numbers = [1, 2, 3, 4, 5]
result = sum_numbers(numbers)
print("Sum of numbers:", result)
"""
    },
    stream=True,
    max_tokens=4096
)

for chunk in response:
    if chunk.choices and len(chunk.choices) > 0 and chunk.choices[0].delta.content is not None:
        print(chunk.choices[0].delta.content, end='')

JavaScript

const prefix = `
def sum_numbers(numbers):
    # 如果列表为空，返回 0
    if not numbers:
        return 0
`;

const suffix = `
# 测试
numbers = [1, 2, 3, 4, 5]
result = sum_numbers(numbers)
print("Sum of numbers:", result)
`;

async function main() {
  const response = await fetch("https://api.nuwaapi.com/v1/chat/completions", {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "Authorization": "sk-***"
    },
    body: JSON.stringify({
      model: "claude-sonnet-4-6",
      messages: [{ role: "user", content: "Please write a sum function code" }],
      stream: true,
      max_tokens: 4096,
      prefix,
      suffix
    })
  });

  const reader = response.body.getReader();
  const decoder = new TextDecoder();

  while (true) {
    const { done, value } = await reader.read();
    if (done) break;

    const lines = decoder.decode(value).split("\n").filter(l => l.startsWith("data: "));
    for (const line of lines) {
      const data = line.slice(6);
      if (data === "[DONE]") return;
      try {
        const chunk = JSON.parse(data);
        const content = chunk.choices?.[0]?.delta?.content;
        if (content) process.stdout.write(content);
      } catch {}
    }
  }
}

main().catch(e => console.error("❌ 出错：", e.message));

TypeScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: "sk-***",
  baseURL: "https://api.nuwaapi.com/v1"
});

const prefix = `
def sum_numbers(numbers):
    # 如果列表为空，返回 0
    if not numbers:
        return 0
`;

const suffix = `
# 测试
numbers = [1, 2, 3, 4, 5]
result = sum_numbers(numbers)
print("Sum of numbers:", result)
`;

async function main(): Promise<void> {
  const stream = await client.chat.completions.create({
    model: "claude-sonnet-4-6",
    messages: [{ role: "user", content: "Please write a sum function code" }],
    stream: true,
    max_tokens: 4096,
    ...({ prefix, suffix } as any)
  });

  for await (const chunk of stream) {
    const content = chunk.choices?.[0]?.delta?.content;
    if (content) process.stdout.write(content);
  }
}

main().catch(e => console.error("❌ 出错：", e.message));

Curl

curl https://api.nuwaapi.com/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: sk-***" \
  -d '{
    "model": "claude-sonnet-4-6",
    "stream": true,
    "max_tokens": 4096,
    "messages": [
      {"role": "user", "content": "Please write a sum function code"}
    ],
    "prefix": "\ndef sum_numbers(numbers):\n    # 如果列表为空，返回 0\n    if not numbers:\n        return 0\n",
    "suffix": "\n# 测试\nnumbers = [1, 2, 3, 4, 5]\nresult = sum_numbers(numbers)\nprint(\"Sum of numbers:\", result)\n"
  }'