托管在香港服务器上的两个 OpenAI 兼容网关：chatgpt2api 负责出图与对话，workbuddy2api 提供多模型文本与视觉能力。全部走 HTTPS，按域名分流。
都兼容 OpenAI 接口格式，用同一套 SDK 就能调
以图为主的网关。支持文生图、图生图、对话，也能走 Anthropic 协议接 Claude Code 类客户端。
/v1/messages 兼容以文本为主的网关。50+ 模型，国内与国际线路可切换，支持视觉读图和长上下文。
cn: / global: 前缀选线路wb；API 调用（/v1/*）只用 Bearer Key，不需要 Basic Auth。chatgpt2api 控制台需要它自己的 Key 登录。把 Key 换成你自己的即可
# 文生图 C2A=https://c2a.yurica.top curl $C2A/v1/images/generations \ -H "Authorization: Bearer $C2A_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "gpt-image-2.5-sunburst", "prompt": "a cyberpunk city", "size": "1024x1024" }'
# 流式对话 WB=https://wb.yurica.top curl $WB/v1/chat/completions \ -H "Authorization: Bearer $WB_KEY" \ -H "Content-Type: application/json" \ -d '{ "model": "cn:deepseek-v4-flash", "messages": [ {"role":"user","content":"你好"} ], "stream": true }'
from openai import OpenAI client = OpenAI( base_url="https://wb.yurica.top/v1", api_key="WB_KEY", ) r = client.chat.completions.create( model="cn:deepseek-v4-flash", messages=[{ "role": "user", "content": "你好", }], ) print(r.choices[0].message.content)
# workbuddy2api 状态 curl https://wb.yurica.top/healthz # 返回示例 {"healthy": 10, "total": 10, "service": "workbuddy2api"}