StarAPI 接入文档

curl -X POST https://cloudapis.cloud/v1/chat/completions \
  -H "Authorization: Bearer sk-你的密钥" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "zai-org/GLM-5.3",
    "messages": [{"role": "user", "content": "你好"}]
  }'

# 先安装 SDK:pip install openai
from openai import OpenAI

client = OpenAI(
    api_key="sk-你的密钥",                    # ← 替换成控制台创建的令牌
    base_url="https://cloudapis.cloud/v1"       # ← StarAPI 接入地址,固定不变
)

resp = client.chat.completions.create(
    model="zai-org/GLM-5.3",                 # 任意可用模型,见「可用模型」章节
    messages=[
        {"role": "user", "content": "你好, StarAPI!"}
    ]
)

print(resp.choices[0].message.content)      # 打印模型回复
print(resp.usage.total_tokens)              # 本次消耗的 Token 数

Authorization: Bearer sk-starapi-xxxxxxxx
Content-Type: application/json

from openai import OpenAI

client = OpenAI(
    api_key="sk-你的密钥",
    base_url="https://cloudapis.cloud/v1"
)

resp = client.chat.completions.create(
    model="Qwen/Qwen3.5-35B-A3B",          # 任意可用模型
    messages=[
        {"role": "system", "content": "你是一个专业的技术助理,回答简洁准确"},
        {"role": "user",   "content": "解释一下什么是 API 网关"}
    ],
    temperature=0.7,                     # 越高越发散,越低越稳定(0~2)
    max_tokens=1024                      # 限制回复长度,控制成本
)

print(resp.choices[0].message.content)

history = [
    {"role": "system", "content": "你是一个耐心的编程助教"}
]

while True:
    q = input("你: ")
    if q in ("exit", "quit"):
        break
    history.append({"role": "user", "content": q})

    resp = client.chat.completions.create(
        model="zai-org/GLM-5.3",
        messages=history
    )
    answer = resp.choices[0].message.content
    print("AI:", answer)

    # 关键:把模型回复也存入历史,模型才"记得"上文
    history.append({"role": "assistant", "content": answer})

from openai import OpenAI

client = OpenAI(
    api_key="sk-你的密钥",
    base_url="https://cloudapis.cloud/v1"
)

stream = client.chat.completions.create(
    model="zai-org/GLM-5.3",
    messages=[{"role": "user", "content": "写一首关于星辰的诗"}],
    stream=True                          # ← 只有这里和普通调用不同
)

full = ""
for chunk in stream:
    delta = chunk.choices[0].delta.content
    if delta:                            # 跳过空片段(如角色/思维链块)
        full += delta
        print(delta, end="", flush=True)

print()                                   # 结束换行,full 即完整回复