文档

将 UnityAI 集成到您的应用所需的一切。

快速入门

UnityAI 提供 100% 兼容 OpenAI 的 API。只需替换 base URL 并使用您的 UnityAI API key — 无需修改代码。

Python

from openai import OpenAI

client = OpenAI(
    api_key="sk-your-unirouter-key",
    base_url="https://api.unityai.cc/v1"
)

response = client.chat.completions.create(
    model="gpt-4o",
    messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)

身份验证

所有 API 请求必须在 Authorization header 中包含您的 API key：

Authorization: Bearer sk-your-api-key

请妥善保管您的 API key。切勿将其暴露在客户端代码或公开仓库中。您可以在 API Keys.

对话补全

POST/v1/chat/completions

为给定的对话创建模型响应。完全兼容 OpenAI Chat Completions API。

请求体

参数	类型	描述
model	string	ID of the model to use (e.g., gpt-4o, claude-3-5-sonnet)
messages	array	A list of messages comprising the conversation so far
stream	boolean	If true, partial message deltas will be sent as SSE events
temperature	number	Sampling temperature between 0 and 2 (default: 1)
max_tokens	integer	The maximum number of tokens to generate
top_p	number	Nucleus sampling parameter (default: 1)

流式输出

设置 stream: true:

const stream = await client.chat.completions.create({
  model: "gpt-4o",
  messages: [{ role: "user", content: "Tell me a story" }],
  stream: true,
});

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

模型

GET/v1/models

列出当前可用的模型。

模型 ID	类型	描述
gpt-4o	chat	OpenAI's flagship multimodal model
gpt-4o-mini	chat	Fast and affordable model for focused tasks
gpt-4-turbo	chat	GPT-4 with vision capabilities
o1-mini	reasoning	Reasoning model optimized for STEM tasks
claude-3-5-sonnet-20241022	chat	Anthropic's most capable model
gemini-1.5-pro	chat	Google's multimodal model with 1M context
dall-e-3	image	High-quality image generation model

频率限制

频率限制基于滑动窗口算法，按每个 API key 施加。

免费层

10 RPM

每分钟 10 次请求

付费层

60 RPM

每分钟 60 次请求

错误码

HTTP 状态码	错误	描述
400	Bad Request	Invalid request parameters
401	Unauthorized	Missing or invalid API key
402	Payment Required	Insufficient balance
429	Too Many Requests	Rate limit exceeded
500	Internal Server Error	Upstream API error or server fault
503	Service Unavailable	All upstream providers are unavailable