LLM API
PhoenixAI 提供与 OpenAI 格式兼容的对话接口，用同一个 sk- 模型 Key 即可调用，按各模型的 token 单价计费。PhoenixAI 负责鉴权、计费与限流。
| 项 | 值 |
|---|---|
| Base URL | https://ai.suzatei.com/v1 |
| 鉴权 | Authorization: Bearer sk-... |
| 对话 | POST /v1/chat/completions |
| 模型列表 | GET /v1/models |
先查有哪些模型
可用模型会随上游调整，请以接口返回为准，不要照抄文档里的名字：
curl -s https://ai.suzatei.com/v1/models \
-H "Authorization: Bearer $PHOENIX_MODEL_KEY"下面的 <模型ID> 请替换为返回结果里的 id。
调用示例
cURL
curl https://ai.suzatei.com/v1/chat/completions \
-H "Authorization: Bearer $PHOENIX_MODEL_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "<模型ID>",
"messages": [{"role": "user", "content": "你好"}]
}'Python（官方 openai SDK）
import os
from openai import OpenAI
client = OpenAI(
api_key=os.environ["PHOENIX_MODEL_KEY"],
base_url="https://ai.suzatei.com/v1",
)
resp = client.chat.completions.create(
model="<模型ID>",
messages=[{"role": "user", "content": "你好"}],
)
print(resp.choices[0].message.content)
print(resp.usage) # prompt_tokens / completion_tokens / total_tokensNode.js
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.PHOENIX_MODEL_KEY,
baseURL: "https://ai.suzatei.com/v1",
});
const resp = await client.chat.completions.create({
model: "<模型ID>",
messages: [{ role: "user", content: "你好" }],
});
console.log(resp.choices[0].message.content);流式输出
加上 stream: true，按 OpenAI 的 SSE 格式逐块返回：
for chunk in client.chat.completions.create(
model="<模型ID>",
stream=True,
messages=[{"role": "user", "content": "写一首短诗"}],
):
print(chunk.choices[0].delta.content or "", end="")