Grok 4 API

xAI 提供支持的 Grok 4 LLM 模型完整 API 参考。

模型变体

此 API 支持两个 Grok 4 模型变体：

快速对比

Grok 4 Text to Text

端点

http复制
POST /api/v1/chat/completions

请求参数

必填

消息支持

可选

示例请求

javascript复制
const response = await fetch('https://api.flaq.ai/api/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Accept': 'text/event-stream',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    model: 'grok-4-text-to-text',
    messages: [
      {
        role: 'user',
        content: 'Explain the key differences between REST and GraphQL APIs.'
      }
    ],
    stream: true,
    max_tokens: 2048
  })
});

Grok 4 Image to Text

端点

http复制
POST /api/v1/chat/completions

请求参数

必填

消息支持

可选

示例请求

javascript复制
const response = await fetch('https://api.flaq.ai/api/v1/chat/completions', {
  method: 'POST',
  headers: {
    'Authorization': 'Bearer YOUR_API_KEY',
    'Accept': 'text/event-stream',
    'Content-Type': 'application/json'
  },
  body: JSON.stringify({
    model: 'grok-4-image-to-text',
    messages: [
      {
        role: 'user',
        content: [
          { type: 'text', text: 'Describe the image and extract any visible text.' },
          {
            type: 'image_url',
            image_url: {
              url: 'https://example.com/sample-image.jpg'
            }
          }
        ]
      }
    ],
    stream: true,
    max_tokens: 2048
  })
});

响应格式

Grok 4 LLM 模型返回 OpenAI 兼容的 completion 响应。使用 stream: true 时，响应为 Server-Sent Events；使用 stream: false 时，响应为单个 JSON 对象。

成功响应

text复制
data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":1710000000,"model":"grok-4-text-to-text","choices":[{"index":0,"delta":{"role":"assistant"},"finish_reason":null}]}

data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":1710000000,"model":"grok-4-text-to-text","choices":[{"index":0,"delta":{"content":"Here is a concise explanation"},"finish_reason":null}]}

data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":1710000000,"model":"grok-4-text-to-text","choices":[{"index":0,"delta":{},"finish_reason":"stop"}]}

data: {"id":"chatcmpl-...","object":"chat.completion.chunk","created":1710000000,"model":"grok-4-text-to-text","choices":[],"usage":{"prompt_tokens":12,"completion_tokens":8,"total_tokens":20}}

data: [DONE]

错误响应

text复制
event: error
data: {"error":{"message":"API requests too frequent, exceeding rate limit","type":"rate_limit_error","code":"1302","param":null}}

最佳实践

1. 使用结构化消息： 通过 messages[] 发送对话历史，而不是把上下文压平成一个 prompt。
2. 选择合适的模型变体： 纯文本推理用 Text to Text，需要图像理解时用 Image to Text。
3. 消费 SSE 事件： 流式展示时追加 choices[0].delta.content，并将 data: [DONE] 视为成功完成。
4. 仅发送支持的输入： Text to Text 不支持图像或文件附件。