Chonway API

Multi-model · Reliable · Cost-controlled

Chonway API: the unified API layer for delivery-focused AI teams

Built for AI development studios, AI SaaS startups, and enterprise AI integration teams. One Base URL to call Claude, GPT and more — reliable, cost-transparent, compatible with your whole toolchain.

Sign up free Enterprise plans
  • Unified multi-model access
  • Dual-protocol
  • Pay-as-you-go
  • Cost-transparent
  • Plug-and-play

Which team type are you?

Different teams have different AI API needs — find your entry point

  • AI Development Studios

    Parallel delivery across multiple projects, frequent model switching — you need reliable access, transparent billing and fast onboarding.

    Read the docs

  • Enterprise / SaaS Teams

    Unified billing management, shared quota across members, stable SLA and high-volume integration support.

    Enterprise plans

  • Content & E-commerce Teams

    Batch-call Claude / GPT for copywriting, translation and SEO content — pay-as-you-go beats a fixed subscription.

    Contact us
Model Matrix

Model Matrix

One entry point covering mainstream vendors and models, with more added over time

  • Anthropic
    • Claude Opus 4.8
    • Claude Sonnet 4.6
    • Claude Haiku 4.5
    • Claude Opus 4.1
    • Claude 3.7 Sonnet
  • OpenAI
    • GPT-5.5
    • GPT-5.4
    • GPT-5.4 mini
    • GPT-4.1
    • o3
  • Google Gemini
    • Gemini 2.5 Pro
    • Gemini 2.5 Flash
    • Gemini 2.0 Flash
  • DeepSeek
    • DeepSeek-V3
    • DeepSeek-R1
  • Alibaba Qwen 通义
    • Qwen3-Max
    • Qwen3-Plus
    • Qwen2.5-72B
  • Baidu ERNIE 文心
    • ERNIE 4.5
    • ERNIE 4.0 Turbo
  • Moonshot Kimi
    • Kimi K2
    • Moonshot v1
  • Tencent Hunyuan 混元
    • Hunyuan-Large
    • Hunyuan-Turbo

The above illustrates the models the platform covers. See live, callable models and real-time prices in the console.

One entry point, fanning out to many models

Your app connects to a single Base URL; Chonway API routes to each vendor’s models

Your App → Chonway API →

  • Anthropic
  • OpenAI
  • Gemini
  • DeepSeek
  • Qwen
  • ERNIE
  • Kimi
  • Hunyuan

Popular & Available

The same Anthropic and OpenAI models, at discounted rates

  • Anthropic 30% OFF

    Claude Opus 4.8

    claude-opus-4-8

    Official $30
    $21 / 1M tokens

    You pay · Save 30%

  • Anthropic 30% OFF

    Claude Sonnet 4.6

    claude-sonnet-4-6

    Official $6
    $4.20 / 1M tokens

    You pay · Save 30%

  • Anthropic 30% OFF

    Claude Haiku 4.5

    claude-haiku-4-5

    Official $1.50
    $1.05 / 1M tokens

    You pay · Save 30%

  • OpenAI 30% OFF

    GPT-5.4

    gpt-5.4

    Official $12
    $8.40 / 1M tokens

    You pay · Save 30%

  • OpenAI 30% OFF

    GPT-5.5

    gpt-5.5

    Official $20
    $14 / 1M tokens

    You pay · Save 30%

  • OpenAI 30% OFF

    GPT-5.4 mini

    gpt-5.4-mini

    Official $0.80
    $0.56 / 1M tokens

    You pay · Save 30%

See live, callable models and real-time prices in the console.

Sign up free View in console

* Prices are illustrative; final billing follows the live console rates.

Six Core Capabilities

Only what the platform genuinely does

  • Request-level failover

    A failing request can automatically switch and retry, so a single point of failure doesn’t break your call.

  • Native dual-protocol

    Native support for both the Anthropic and OpenAI protocols — one key, two ecosystems.

  • Real model names

    Call models by their real names — no renaming, no obfuscation. What you see is what you call.

  • Transparent billing

    Billed per token, anchored to official pricing, no monthly fee. Every call is auditable in the console.

  • Cheaper than official

    Mainstream models start at up to 30% off official pricing — the same models, for less.

  • Plug-and-play

    Claude Code, Codex and mainstream SDKs work by swapping the Base URL — no business-code changes.

Pricing Advantage

The same models, for less

  • Up to 30% off official

    Mainstream Claude and GPT models start at 30% off — the more you call, the more you save.

  • Pay-as-you-go, no monthly fee

    No plan lock-in, no minimum top-up. You pay only for what you use, and credits never expire.

  • Transparent billing

    The console logs every call — model, tokens and cost — so your bill is always crystal clear.

Integration

Change one Base URL, integrate in three steps, no business-code rewrites

Unified Base URL https://api.wos.one/v1

  1. Claude Code

    Set the environment variables to use the full Claude lineup inside Claude Code — no other config needed.

    export ANTHROPIC_BASE_URL=https://api.wos.one
export ANTHROPIC_API_KEY=your-key
claude

  2. Codex / OpenAI-compatible

    Set the OpenAI Base URL to the Chonway API endpoint and tools like Codex and Cursor work out of the box.

    export OPENAI_BASE_URL=https://api.wos.one/v1
export OPENAI_API_KEY=your-key

  3. SDK

    With the official SDKs, just swap base_url — the rest of your code stays untouched.

    from openai import OpenAI
client = OpenAI(
    base_url="https://api.wos.one/v1",
    api_key="your-key",
)

Why Chonway API

We handle reliability so you can focus on your product

  • Reliable

    Request-level failover and retry, so a single point of failure won’t break your call.

  • Dual-protocol

    Compatible with both Anthropic and OpenAI API protocols — one key, two ecosystems.

  • Transparent billing

    Settled per call against official metering. No hidden fees; every charge is traceable.

  • Developer-friendly

    Works with mainstream CLIs and SDKs. Clear docs let you migrate in minutes.

Use Cases

A fit for individuals and teams alike

  • Individual developers

    Call top-tier Claude and GPT models affordably — side projects, coding and experiments without the sticker shock.

  • Heavy CLI users

    Code all day with Claude Code and Codex. Pay-as-you-go is cheaper and more controllable than subscriptions.

  • Enterprise teams

    Unified keys and billing, shared quota across the team, transparent costs for easy accounting.

FAQ

What you may want to know about setup, billing and reliability

Is Chonway API compatible with the official API?

Fully compatible. We support both the Anthropic and OpenAI API protocols. Just point the Base URL to https://api.wos.one/v1 (use https://api.wos.one for the Anthropic protocol) — the request format matches the official API, so no business-code changes are needed.

How do I set up Claude Code?

Set ANTHROPIC_BASE_URL=https://api.wos.one and ANTHROPIC_API_KEY=your-key, then run claude to use the full Claude lineup.

How does billing work?

Pay-as-you-go — you pay only for what you use, with no monthly fee or plan lock-in. The console logs each call (model, tokens, cost) for full transparency, and credits never expire.

How much cheaper is it than official pricing?

Mainstream Claude and GPT models start at up to 30% off official pricing — the more you call, the more you save. Exact rates follow the live console pricing.

How reliable is the service?

We run a multi-node high-availability architecture with automatic failover and load balancing, plus routing optimized for low latency and high uptime.

How do I sign up and log in?

Click “Sign up free” to create an account at https://console.wos.one/register; existing users can log in directly at https://console.wos.one.

Learn

Guides, comparisons & answers

From integration guides to model picks — get more out of the full Claude & GPT lineup, faster

Hand AI integration over to Chonway API today

One Base URL for the full Claude & GPT lineup. Instant access, pay-as-you-go, up to 30% off official.

Sign up free Read the docs