一个访问令牌,同时使用 GPT 与 Claude 全系列模型。按下面的步骤,几分钟内就能接好。 One access token, full access to both the GPT and Claude model families. Follow the steps below — it only takes a few minutes.
https://api.vocentraai.com
Note the gateway address: https://api.vocentraai.com
如果还没收到令牌,请先通过购买渠道联系我们完成开通,再回来看下面的步骤。 If you haven't received a token yet, please contact us through the channel you purchased from first, then come back to the steps below.
适合:在终端里用 Claude Code 写代码、跑任务。 Best for: coding and running tasks with Claude Code in your terminal.
export ANTHROPIC_BASE_URL="https://api.vocentraai.com"
export ANTHROPIC_API_KEY="YOUR_TOKEN"
Windows(PowerShell): Windows (PowerShell):
$env:ANTHROPIC_BASE_URL="https://api.vocentraai.com"
$env:ANTHROPIC_API_KEY="YOUR_TOKEN"
claude
提示:以上设置只在当前终端窗口生效。想让它每次都自动生效,把这两行 export 加到 ~/.zshrc 或 ~/.bashrc 文件末尾(Windows 可加到 PowerShell 配置文件,或在系统环境变量里设置)。想固定使用某个模型,可以再加一行 export ANTHROPIC_MODEL="claude-sonnet-5"。
Tip: this only applies to the current terminal session. To make it permanent, add the two export lines to the end of your ~/.zshrc or ~/.bashrc (on Windows, add them to your PowerShell profile or set them as system environment variables). To pin a specific model, add export ANTHROPIC_MODEL="claude-sonnet-5".
适合:Cherry Studio、LobeChat、NextChat 等支持"自定义 API"的聊天客户端,以及自己写代码调用。 Best for: chat clients that support a custom API — Cherry Studio, LobeChat, NextChat, etc. — and your own code.
如果客户端要求的地址不带 /v1,就只填 https://api.vocentraai.com;API Key 不需要手动加 "Bearer" 前缀,客户端会自己处理。
If the client asks for a URL without /v1, use https://api.vocentraai.com instead. No need to prefix the key with "Bearer" — the client handles that.
命令行测试(把 YOUR_TOKEN 换成你的令牌):
Test from the command line (replace YOUR_TOKEN with your token):
curl https://api.vocentraai.com/v1/chat/completions \
-H "Authorization: Bearer YOUR_TOKEN" \
-H "Content-Type: application/json" \
-d '{"model":"gpt-4o","messages":[{"role":"user","content":"Hello"}]}'
Python (openai):
from openai import OpenAI
client = OpenAI(
api_key="YOUR_TOKEN",
base_url="https://api.vocentraai.com/v1",
)
resp = client.chat.completions.create(
model="gpt-4o",
messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)
适合:需要原生 Anthropic Messages API 格式的 SDK 或工具。 Best for: SDKs or tools that expect the native Anthropic Messages API format.
命令行测试:Test from the command line:
curl https://api.vocentraai.com/v1/messages \
-H "x-api-key: YOUR_TOKEN" \
-H "anthropic-version: 2023-06-01" \
-H "Content-Type: application/json" \
-d '{"model":"claude-sonnet-5","max_tokens":100,"messages":[{"role":"user","content":"Hello"}]}'
Python (anthropic):
import anthropic
client = anthropic.Anthropic(
api_key="YOUR_TOKEN",
base_url="https://api.vocentraai.com",
)
msg = client.messages.create(
model="claude-sonnet-5",
max_tokens=100,
messages=[{"role": "user", "content": "Hello"}],
)
print(msg.content)
具体可用范围以你的套餐为准;下面是当前网关上线的模型。 Your exact access depends on your plan. Below is the current list live on the gateway.
| 系列Family | 模型名Model name | 调用格式Format |
|---|---|---|
| GPT | gpt-4o-mini | OpenAI |
| GPT | gpt-4o | OpenAI |
| GPT | gpt-5.6-sol | OpenAI |
| GPT | gpt-5.6-terra | OpenAI |
| GPT | gpt-5.6-luna | OpenAI |
| GPT | gpt-6-astra | OpenAI |
| Claude | claude-haiku-4-5-20251001 | Anthropic |
| Claude | claude-sonnet-5 | Anthropic |
| Claude | claude-opus-5 | Anthropic |
令牌复制错误或包含多余空格。建议逐字核对,或联系我们获取一份新的。 The token was mistyped or has extra whitespace. Check it character by character, or contact us for a fresh copy.
模型名拼错,或该模型不在你的套餐内 —— 对照上方"可用模型"列表检查。 The model name is misspelled, or isn't included in your plan — check it against the "Available Models" table above.
额度用完了,需要续费或充值,请联系我们。 You're out of quota — contact us to top up or renew.
确认用的是 https 而不是 http,且域名没有拼错。
Make sure you're using https (not http) and the domain is spelled correctly.
通过你购买时使用的联系方式联系我们,我们会尽快处理。 Reach us through the same channel you used to purchase — we'll get back to you as soon as we can.