1xToken1xToken

Self-hosted AI Gateway

Live

One endpoint, every model

Self-hosted AI gateway. Swap the API URL and you're in

https://api.1xtoken.cn/v1
Get API KeyRead docs
Chat Completions
curl https://api.1xtoken.cn/v1/chat/completions \
  -H "Authorization: Bearer $ONE_X_TOKEN_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpt-4.1",
    "messages": [
      { "role": "user", "content": "你好" }
    ]
  }'

5+

Model families

1

Unified API key

OpenAI

Compatible API

End-to-end

Usage & billing

Core Features

Bring model access, routing, and cost management into one simple, reliable infrastructure layer.

One interface for major models

Compatible with mainstream SDKs and request formats, so you can switch model providers without rewriting application code.

Flexible routing

Choose the right model based on performance, price, and availability so each request can take a better path.

Clear cost visibility

Review usage and spending in one place, and manage team keys and model budgets together.

Supported models

Not tied to a single upstream. Major model families are already adapted. Actual availability depends on admin-configured channels and upstreams.

OpenAIChatGPT
ClaudeClaude
DeepSeekDeepSeek
MinimaxMiniMax
ZhipuGLM
GeminiGemini
QwenQwen
GrokGrok

One endpoint, routed to every model

Connect once, route to any model on demand

Same Base URL and key. Requests route to the matching upstream by model name, with no backend changes in your code.

Base URL
api.1xtoken.cn/v1
glm-5.2model
deepseek-v3model
minimax-m3model

Workflow

Keep your existing development habits and connect model requests to the unified gateway in three steps.

01

Create a key

Generate an API Key in the console and manage access separately for different projects.

02

Replace the URL

api.openai.com/v1api.1xtoken.cn/v1

03

Send requests

Choose the model you need and review all usage from one account.

Governance & operations

Calling models is just the start. 1xToken folds user groups, quota limits, call logs, and risk rules into one governable backend, so every team manages model services by its own operating rules, not just an API address.

Users & quotas

Manage users, groups, and quotas by your own operating rules, with independent access and budgets per project.

Billing & invoices

Keep costs transparent through multipliers, groups, and logs, with consolidated spend and invoices.

Logs & audit

Full call logs and quota changes are traceable, so every request and quota adjustment is accountable.

Risk & notices

Channel risk control, usage alerts, and in-site announcements keep operations visible to admins.

Why you need it

Connecting to multiple model providers means managing many keys, many bills, and each one's availability swings. 1xToken folds all of that into one entry point: one URL, one quota, one log, so teams stop spreading thin on multi-source model ops.

Ready to call every model from one address?

Get startedRead the docs