API 接口文档

图像/视频生成

提交图像或视频生成请求，支持 GPT Image、Nano Banana、Veo3 等模型。异步接口，返回任务 ID 后需轮询查询结果。

请求参数

请求示例

# cURL
curl -X POST https://your-domain.com/api/v1/generate \
  -H "Authorization: Bearer sk-grsai-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2-vip",
    "prompt": "A cute cat wearing sunglasses",
    "aspectRatio": "1:1"
  }'

响应示例

{
  "code": 0,
  "message": "操作成功",
  "data": {
    "id": "task-abc123",
    "status": "succeeded",
    "results": [
      { "url": "https://cdn.grsai.com/generated/image.jpg" }
    ]
  }
}

查询异步结果

生成接口返回任务 ID 后，通过此接口轮询查询生成结果。建议每 2-3 秒轮询一次。

请求示例

curl -X GET "https://your-domain.com/api/v1/result?id=task-abc123" \
  -H "Authorization: Bearer sk-grsai-xxx"

响应示例

{
  "code": 0,
  "data": {
    "id": "task-abc123",
    "status": "succeeded",
    "results": [{ "url": "https://cdn.grsai.com/result.jpg" }]
  }
}

对话补全（OpenAI 兼容）

支持 GPT-5.4、Gemini 等文本对话模型，接口格式兼容 OpenAI Chat Completions API。支持流式（SSE）和非流式两种模式。

请求参数

请求示例（非流式）

curl -X POST https://your-domain.com/api/v1/chat/completions \
  -H "Authorization: Bearer sk-grsai-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4",
    "messages": [
      { "role": "user", "content": "Hello, who are you?" }
    ]
  }'

响应示例（非流式）

{
  "code": 0,
  "data": {
    "id": "chat-xxx",
    "choices": [{
      "message": { "role": "assistant", "content": "I am an AI assistant..." }
    }],
    "usage": { "prompt_tokens": 10, "completion_tokens": 20 }
  }
}

请求示例（流式）

curl -X POST https://your-domain.com/api/v1/chat/completions \
  -H "Authorization: Bearer sk-grsai-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-5.4",
    "messages": [
      { "role": "user", "content": "Hello!" }
    ],
    "stream": true
  }'

响应示例（流式 SSE）

# 响应头: Content-Type: text/event-stream
data: {"id":"chat-xxx","choices":[{"delta":{"content":"Hello"}}]}

data: {"id":"chat-xxx","choices":[{"delta":{"content":"!"}}]}

data: [DONE]

图片生成（OpenAI 兼容）

兼容 OpenAI Images API 格式的图片生成接口。

请求示例

curl -X POST https://your-domain.com/api/v1/images/generations \
  -H "Authorization: Bearer sk-grsai-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nano-banana-pro",
    "prompt": "A sunset over mountains",
    "size": "1024x1024",
    "response_format": "url"
  }'

批量生成

一次请求生成多张图片（1-10张），每张独立计费。

请求示例

curl -X POST https://your-domain.com/api/v1/batch-generate \
  -H "Authorization: Bearer sk-grsai-xxx" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-image-2",
    "prompt": "A beautiful landscape",
    "count": 3,
    "aspectRatio": "16:9"
  }'

模型列表

获取所有可用模型及其定价信息。

curl -X GET https://your-domain.com/api/v1/models \
  -H "Authorization: Bearer sk-grsai-xxx"

响应示例

{
  "object": "list",
  "data": [
    {
      "id": "gpt-image-2",
      "display_name": "GPT Image 2",
      "category": "image",
      "pricing_type": "per_request",
      "cost_credits": 600,
      "description": "高质量图像生成模型"
    },
    {
      "id": "rh-video",
      "display_name": "RunningHub 图生视频",
      "category": "video",
      "pricing_type": "per_request",
      "cost_credits": 3000,
      "description": "图生视频模型，支持时长6-30秒"
    }
  ]
}

图像模型

视频模型

注意：视频模型调用与图像模型使用相同的 API 端点（/api/v1/generate、/api/v1/chat/completions 等），系统会根据模型 category 自动路由到对应上游供应商。

视频生成额外参数：resolution（分辨率）、duration（时长秒数）、aspectRatio（16:9 / 9:16）。

账户信息

查询当前 API Key 关联的账户信息及积分余额。

curl -X GET https://your-domain.com/api/v1/me \
  -H "Authorization: Bearer sk-grsai-xxx"

响应示例

{
  "code": 0,
  "data": {
    "id": "user-xxx",
    "email": "user@example.com",
    "credits": 4800,
    "balance": 0,
    "plan": "free"
  }
}