bn-router
One endpoint for every model

Route all your AI APIs through one gateway

bnrouter lets you use Claude, GPT, Gemini, Kimi, DeepSeek, Qwen, Grok, Llama, Flux and more behind a single OpenAI-compatible endpoint. Create keys, set quotas, monitor usage, and pay as you go.

Supports major model families

ClaudeGPTGeminiKimiDeepSeekQwenGrokLlamaFlux
Get startedRead the docs
Auth
Quota
Route
Meter

route manifest

POST /v1/chat/completions

live
authkey ok
quotagroup A
policybest fit
meterlive cost

selected lane

42ms
chatactive
embedstandby
imagestandby

1

Unified endpoint

Multi

Model providers

3

Interface languages

Everything you need to ship AI features

A production-grade gateway for teams that call many models and want one control plane.

control plane
healthy
keys24
quota68%
spend$0.41
cachehot
auth
route
meter

  1. One compatible API

    Use Claude, GPT, Gemini, Kimi, DeepSeek, Qwen and more with the OpenAI chat & embeddings protocol. Swap models without rewriting client code.

  2. Keys, groups & quotas

    Issue per-app API keys, assign them to rate-limited groups, and cap daily, weekly, or monthly usage.

  3. Usage & cost visibility

    Track requests, tokens, and cost per key, model, and platform in real time from a clean dashboard.

  4. Subscriptions & top-ups

    Sell quota-based subscription plans or balance top-ups with third-party payment hosting and redeem codes.

Three steps from signup to your first streaming response.

  1. 1account

    Create an account

    Register in seconds, then top up your balance or pick a subscription plan.

  2. 2key

    Create an API key

    Issue a key, choose a group, and set usage limits from the console.

  3. 3request

    Point your app at us

    Set the base URL to our endpoint and call any model with your existing client SDK.

Built on your gateway

bnrouter does not just route models — it runs agents on top of them. Each agent turns a plain-language brief into a finished deliverable, using one key.

Agent case 01

Ad Landing Page Agent

Describe your product in one sentence. The agent writes the copy, builds A/B variants in up to three languages, embeds tracking pixels, and publishes a hosted ad landing page with SEO files — ready for Google Ads, Meta, and TikTok.

Landing pageA/B variants3 languagesTracking pixelsSEO files
Try in console
bn agent · ad-landingLive
$bn agent --ad-landing
›brief: "Acme AI — ship faster"
›copy: hero A/B · faq · 3 langs
›embed: gtag · meta · tiktok
✓published · 5 files · /landings/acme-ai
$
Image & video generation

Generate production-ready visuals through the same unified endpoint — no separate accounts or integrations needed.

6+ models

Image generation

Produce high-quality images from text prompts or edit existing ones. Works with leading image models including Flux, DALL·E, Stable Diffusion, and more — all through a single OpenAI-compatible endpoint.

Flux ProDALL·E 3Stable Diffusion 3Midjourney (via proxy)FLUX.1 DevFLUX.1 Schnell
3+ models

Video generation

Generate short-form video clips from text or image prompts. Support for emerging video models including Sora, Kling, and others — expanding rapidly as new capabilities become available.

Sora (OpenAI)Kling (Kuaishou)Runway Gen-3
bn route · imagesLive
$bn route --image --model flux-pro
›POST /v1/images/generations
›prompt: "A serene mountain lake at sunrise"
✓200 OK · 3 image(s) · url[0]
$

Explore the API reference

The endpoint follows the OpenAI images format — send a POST to /v1/images/generations with your prompt, and the gateway routes to the best available upstream account.

Image & video API docs

The endpoint speaks the OpenAI protocol, so any client that works with OpenAI works here — just swap the base URL and your key.

Integration steps

  1. 1

    Create an account

    Sign up in seconds and add a small balance.

  2. 2

    Create an API key

    Issue a key from the console and set your usage limits.

  3. 3

    Make a request

    Point your OpenAI-compatible client at our endpoint. No SDK changes needed.

Example request
curl https://www.bnrouter.com/v1/chat/completions \
  -H "Authorization: Bearer $YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "your-model-id",
    "messages": [{"role": "user", "content": "Hello!"}]
  }'

Replace YOUR_API_KEY with a key from the console. The exact base URL is shown in your dashboard.

Quick answers about what bnrouter is and how it works.

What is bnrouter?

bnrouter is an AI API gateway. It aggregates multiple large-model providers behind a single OpenAI-compatible endpoint, so your application talks to one API instead of many.

How is it different from calling providers directly?

Instead of managing separate accounts, SDKs, and billing for every provider, you get one key, one endpoint, unified quotas, and one usage dashboard — while models remain interchangeable.

Is it free to use?

The gateway itself has no setup fee. You only pay for the model usage you consume, either via balance top-ups or quota-based subscription plans.

Is my data safe?

Yes. Requests are forwarded to the model providers over encrypted channels. We do not store your prompts or completions; only usage metadata is kept for billing and analytics.

Quick answers about what bnrouter is and how it works.
Offers & support

Seedance unlimited, Claude & GPT discounts

Talk to us on Telegram for onboarding, quotes, and integration help.

Video

Seedance unlimited API

On-demand video generation without call-count quotas — built for batch and long-running workloads.

Chat

Claude / GPT low discounts

Major chat models at discounted rates so you can keep inference costs under control.

Contact support on Telegram

Ask about pricing, keys, routing, or troubleshooting. Include your account email for faster help.

@pupuxiaoyang

Open Telegram

Start routing your AI traffic today

Create an account and make your first request in under five minutes.

Open the console