Access 200+ Models · One API Key

All Your AI Models,
One Endpoint. Zero Friction.

One API key unlocks OpenAI, Anthropic, Google, Meta, DeepSeek, and 50+ other providers —all behind a single OpenAI-compatible endpoint. No more juggling eight different billing accounts, wrestling with provider-specific SDKs, or worrying about regional blocks. Ship AI features, not infrastructure.

Get StartedSee All Models
Powered by Leading AI Labs

Your single endpoint connects to every frontier model on the market —zero integration effort.

OpenAIAnthropicGoogleMetaDeepSeekMistral+200 models →
200 ok
POST/v1/chat/completions
Request
curl -X POST "/v1/chat/completions" \
-H "Authorization: Bearer sk-•••• \
-d '{
"model": "your-model",
"messages": [
{ "role": "user", "content": "..." }
]
}'
Response
{
"choices": [{ "message": { "content": "Hello! How can I help you today?" } }],
"usage": { "total_tokens": 27 }
}
Request
curl -X POST "/v1/responses" \
-H "Authorization: Bearer sk-•••• \
-d '{
"model": "your-model",
"input": "..."
}'
Response
{
"output": [{ "content": "Here is the response you requested." }],
"usage": { "total_tokens": 35 }
}
Request
curl -X POST "/v1/messages" \
-H "x-api-key: sk-ant-•••• \
-d '{
"model": "claude-sonnet-4-6",
"max_tokens": 1024,
"messages": [
{ "role": "user", "content": "..." }
]
}'
Response
{
"content": [{ "text": "I appreciate your question —let me help with that." }],
"usage": { "input_tokens": 12, "output_tokens": 8 }
}
Request
curl -X POST "/v1/models/gemini-pro:generateContent" \
-H "x-goog-api-key: AIza•••• \
-d '{
"contents": [
{ "parts": [{ "text": "..." }]
} ]
}'
Response
{
"candidates": [{ "content": { "Response generated successfully." } }],
"usageMetadata": { "totalTokenCount": 18 }
}
142ms27tokenscost $0.00081
168ms35tokenscost $0.00120
205ms20tokenscost $0.00150
189ms18tokenscost $0.00095
stream · sse
50+AI Providers Integrated
200+Models Available
99.99%API Uptime SLA
<1msRouting Overhead

Why OneRouter

One Integration. Every Model. Zero Lock-In.

Stop managing a patchwork of AI provider accounts. OneRouter replaces the complexity with a single, elegant API layer —so your team ships features instead of infrastructure.

01

Zero Provider Lock-In

Switch between OpenAI, Claude, Gemini, and DeepSeek with a single base URL change. No SDK rewrites, no pipeline rework —your code stays the same, the model changes instantly.

OpenAI
Anthropic
Google
Meta
DeepSeek
xAI
02

Global API Access

Route through 15+ edge locations to bypass regional blocks. Your developers in Asia, Europe, and the Americas all get sub-50ms latency.

03

Transparent Pricing

Small, transparent service fee on top of provider rates. No hidden costs, no minimum commitment, no surprise bills. Full cost visibility on every request.

1
Provider Pricing
2
Service Fee
3
Real-Time Tracking
04

Drop-In OpenAI Replacement

Any library that speaks the OpenAI API now speaks 200+ models. Python, Node.js, Go, curl —change one environment variable, unlock every model on the market.

Python
Node
Go
curl
100% OpenAI-Compatible

Auto Failover

If a provider goes down, traffic auto-routes to your backup model —no dropped requests

Cost Optimization

Set per-token budgets, spending caps, and smart routing to the cheapest equivalent model

Team Workspaces

Multi-user accounts with RBAC, shared billing, and per-project API key scoping

Open Source Core

AGPL-3.0 open-source, self-hostable, 39K+ GitHub stars, 50+ community contributors

Use Cases

Built for Builders, Resellers, and Enterprises

Whether you're shipping an AI product, reselling API access, or deploying behind your firewall —OneRouter adapts to your model.

AI Startups & Scale-Ups

Launch AI features in hours, not months. One integration gives your product access to every major model on the market.

API Resellers & White-Label

Rebrand and resell AI API access under your own domain with custom pricing, usage limits, and branded dashboards.

AI Agents & Chatbots

Power your agents with dynamic model routing —automatic fallback, A/B testing, and cost-optimized dispatch logic.

Enterprise Teams

Centralized billing, SAML SSO, and department-level API key scoping. Manage AI spend across your entire org.

Content & Media Platforms

Text generation, image understanding, audio transcription, video analysis —one unified pipeline, zero integration overhead.

Self-Hosted / On-Premise

Full data sovereignty behind your firewall. Deploy via Docker, Kubernetes, or bare metal with complete audit control.

Model Matrix

Comprehensive Coverage Across Every AI Provider

One API key unlocks every major model. Switch providers without changing a single line of code.

Large Language Models

GPT-4o
GPT-4.1
Claude Opus 4.8
Claude Sonnet 4.6
Gemini 2.5 Pro
Gemini 2.5 Flash
DeepSeek V3
DeepSeek R1
Qwen3
Llama 4
Mistral Large 3
More Models…

Vision & Multimodal

GPT-4o Vision
Claude 4 Vision
Gemini 2.5 Vision
Qwen-VL
Llama 4 Vision
More Models…

Embeddings & Reranking

text-embedding-3
Cohere Embed
Voyage AI
BGE-M3
Jina Reranker
More Models…

Speech & Audio

GPT-4o TTS
Whisper
Gemini TTS
ElevenLabs
Fish Audio
More Models…

Compliance & Security

Enterprise-Grade Protection, Global Standards

Bank-level encryption, granular access controls, and full audit trails —because your data deserves nothing less.

End-to-End Encryption

All data in transit is protected with TLS 1.3 and at rest with AES-256. Zero-trust architecture by default —your keys, your control.

TLS 1.3AES-256

Access Control

Granular role-based access with SSO/SAML, API key scoping, IP whitelisting, and mandatory MFA for all admin accounts.

RBACSSO

Audit Logging

Comprehensive audit trails for every API call, config change, and admin action. SIEM-ready export formats with real-time streaming.

SIEMLog Auditing
Certifications & StandardsSOC 2 Type IIGDPRISO 27001HIPAACCPA

Trusted by Builders Worldwide

500M+ Tokens Routed Daily Across 50 Countries

From indie developers to enterprise teams —OneRouter is the open-source backbone powering AI API infrastructure globally.

5.2K

GitHub Stars

50K+

Docker Pulls

50+

Open-Source Contributors

500M+Tokens Routed Daily
10,000+Active Developers
500+Teams & Enterprises
50+Countries Served

Quick Start

From Zero to Production in 5 Minutes

Already have an OpenAI key? You're 30 seconds away from unlocking every model on the market.

1

Import Your Keys

Already have OpenAI, Anthropic, or Google keys? Add them in 30 seconds. No new signups needed.

# Add your existing provider keys
openai:
  key: sk-your-openai-key
anthropic:
  key: sk-ant-your-key
2

Swap Your Base URL

Change one environment variable in your app. Every OpenAI-compatible library now speaks 200+ models.

# Before: locked into one provider
OPENAI_BASE_URL="https://api.openai.com/v1"

# After: every model, one endpoint
OPENAI_BASE_URL="https://api.onerouter.com/v1"
3

Ship Your AI Feature

That's it. Your app now routes through OneRouter. Monitor usage, set budgets, and scale —all from one dashboard.

$ curl -X POST https://api.onerouter.com/v1/chat/completions \
  -H "Authorization: Bearer sk-your-key" \
  -d '{ "model": "gpt-4o", "messages": [...] }'
Read the Full Docs

One API Key. 200+ Models.
Start in 5 Minutes.

Pay only for the tokens you use. No subscription, no minimum commitment. Top up and start calling 200+ models in minutes.

Get StartedTalk to Sales

FAQ

Still Have Questions?

From data privacy to failover —everything you need to know before you deploy.

What is OneRouter and how does it work?

OneRouter is a unified AI API gateway that provides a single OpenAI-compatible endpoint to access 200+ models across providers including OpenAI, Anthropic, Google, Meta, and DeepSeek. It handles routing, load balancing, rate limiting, cost tracking, and automatic failover —so you can focus on building instead of managing infrastructure.

Is OneRouter compatible with existing OpenAI SDKs and tools?

Yes, absolutely. OneRouter exposes a fully OpenAI-compatible API —any library, SDK, or tool that works with the OpenAI API works with OneRouter out of the box. Just change the base URL to your OneRouter endpoint and you instantly gain access to every supported model across every provider. No code changes required.

How does pricing work? Is there a free tier?

OneRouter charges a small, transparent service fee on top of provider rates — no hidden costs, no minimum commitments. Our pricing model ensures you get reliable infrastructure, global access, and 24/7 support without unpredictable bills. The open-source core (AGPL-3.0) is freely available for self-hosted deployments.

Can I self-host OneRouter on my own infrastructure?

Yes. OneRouter is fully self-hostable via Docker, Docker Compose, or Kubernetes. We provide comprehensive deployment guides, a Helm chart for production Kubernetes clusters, and bare-metal installation scripts. The self-hosted version includes every feature with no usage limits —ideal for enterprises with data residency or compliance requirements.

Do you store or log my API requests and responses? How is failover handled?

No. We never store your prompts, completions, or embeddings. OneRouter proxies requests in real-time —your data passes through, it doesn't stay. API keys are encrypted at rest (AES-256). For failover: if an upstream provider goes down, OneRouter automatically routes to your next preferred model based on your fallback chain. Zero dropped requests, zero manual intervention. Self-hosted deployments give you complete data sovereignty under your existing infrastructure policies.