One Key, Access Multiple Leading Models

Unified OpenAI-compatible API with auto-routing to OpenAI, Anthropic, DeepSeek, Qwen and more. Built-in balance management, model-level billing, failover and usage tracking.

OpenAI-Compatible Auto Health Checks Usage-Based Billing UKey + DKey

curl -X POST /v1/chat/completions
  -H "Authorization: Bearer <your-key>"
  -H "Content-Type: application/json"
  -d '{
    "model": "gpt-4o",
    "messages": [{"role":"user","content":"Hello"}],
    "stream": true
  }'

# Auto routing + failover + real-time billing

99.9%Gateway availability target

< 300msAverage routing overhead

10+Mainstream model providers supported

Unified BillingTraceable model-level cost breakdown

An AI Gateway Built for Production

Reduce provider switching, quota management and billing complexity so your team can focus on product logic.

Unified Multi-Model Endpoint

Use a standard OpenAI-style API and switch underlying providers without changing your SDK.

Reliability First

Provider health checks and auto-recovery keep unstable nodes out and restore them safely.

Fine-Grained Billing

Track input/output tokens by model with separate pre-charge and settlement logic.

Multi-Channel Payments

Support Stripe and on-chain payments (ERC-20 / Solana) with automatic status polling.

Dual-Key System

Support user-level UKey and device-level DKey for layered authorization scenarios.

Observability

Built-in reports for usage, costs, popular models and provider health.

Simple and Transparent Pricing

Pay for what you use or prepay with packages. Actual pricing is configured in your console.

Developer

$0 / month

Create API keys right after registration
Usage-based billing for testing and small traffic
Core routing and failover

Create Account

Recommended

Team

Top-up Packages Deduct from balance

Centralized member and API key management
Stripe / Crypto top-up support
Model-level usage and cost tracking

Open Console

Enterprise

Custom Private deployment optional

Environment isolation and permission strategy
Custom billing rules and reports
High-availability deployment guidance

View Architecture Docs

3 Steps to Go Live

STEP 01

Create Account and Generate Key

STEP 02

Switch `base_url`

Point your existing OpenAI SDK `base_url` to `/v1`, then replace API key.

STEP 03

Track Usage and Cost

Monitor model calls, cost trends and provider health in the console.

FAQ

Is it compatible with existing OpenAI SDKs?

Yes. Replace only `base_url` and API key, keeping the existing request shape.

How do I switch between models?

Set the `model` field in request. The gateway maps and routes to healthy providers.

How long does top-up confirmation take?

Stripe is usually confirmed in seconds; on-chain payments update after confirmations.

Cut Integration Complexity to the Minimum

Start today and replace multi-vendor glue code with one unified gateway.

Start Now Browse API Docs