OpenRouter · API platform (model router)

OpenRouter

One OpenAI-compatible API key for hundreds of models from many providers.

LLM APIs & Model Platforms Inference & Model Hosting

Provider: OpenRouter
Latest confirmed model: Not applicable / not verified
Pricing: Free plan + paid
Free plan: Yes
API: Yes · free quota
Status: Generally available
Last verified: Oct 19, 2026

Visit official website Pricing page API docs

What is OpenRouter?

OpenRouter is a single API that forwards your requests to 500+ models hosted by 80+ providers, so you can switch models without changing integrations. It is compatible with the OpenAI SDK, handles fallbacks between providers, and bills you at each provider's list price plus a platform fee when you buy credits.

Managing separate accounts, keys and SDKs for every model provider, and coping with provider outages.

Why use OpenRouter?

Practical benefits

One key and bill for many providers
Drop-in OpenAI SDK compatibility
Automatic provider fallbacks
Free models for experimentation

Who it suits

Developers comparing models
Apps needing redundancy across providers
Hobbyists using free models

When another tool may fit better

Teams that must contract directly with one model vendor
Very high-volume users sensitive to the credit purchase fee
Offline use

Learning curve: Beginner-friendly
Skills needed: Basic API usage.

Example workflows

Point the OpenAI SDK base URL to https://openrouter.ai/api/v1 and change the model slug to compare models

What can it do?

Unified API500+ models from 80+ providers through one endpoint.
Free models Limited25+ free models with daily request caps.
BYOK Depends on planBring your own provider keys; first $25,000 of list-price inference per month carries no OpenRouter fee on Standard/Business, then 5%.
Routing and spend controls Depends on planAuto-routing, budgets, prompt caching and data-policy routing on paid plans.
EU in-region routing Depends on planAvailable on Business and Enterprise.
SDKsTypeScript and Python client SDKs plus an Agent SDK.

Good for

Comparing models quickly
Multi-provider redundancy
Cost-optimised routing
Free experimentation

Where it runs

WebREST API

Pricing and plans

Free plan + paid No subscription: inference is billed at each provider's list price, with a platform fee of 5.5% (Standard) or 8% (Business) on credit purchases. A free plan offers 25+ free models with request caps; Enterprise is contract-based.

Plan	Price	Billing	Free trial / tier	Usage limits	Main inclusions	Model access	API access	Restrictions
Free	Free	usage-based	Free tier	50 requests/day and 20 requests/minute; 1,000/day after buying at least $10 in credits	25+ free models 4 free providers 5 workspaces	Free models	Yes	Failed attempts count toward the cap; negative balance blocks free models
Standard (pay as you go)	Provider list prices plus 5.5% fee on credit purchases	usage-based	—	Passthrough provider limits	500+ models Auto-routing Budgets Prompt caching BYOK	All models	Yes	—
Business (pay as you go)	Provider list prices plus 8% fee on credit purchases	usage-based	—	Passthrough provider limits	1,000 workspaces EU/US in-region routing Signed DPA on request	All models	Yes	Can downgrade 30 days after upgrading
Enterprise	Contact provider	usage-based	—	Higher allowances	SSO/SAML SLAs Invoicing Dedicated support	All models	Yes	—

Prices in USD · Global · Listed prices exclude taxes · Verified Oct 19, 2026

Official pricing page (opens in a new tab)

API and developer access

Official API: Yes
Documentation: API docs (opens in a new tab)
Authentication: API key (Bearer)
Pricing model: per 1M tokens at provider list price plus credit-purchase fee
Free API quota: Free models: 50 requests/day, 20 requests/minute (1,000/day after buying $10+ credits)
Rate limits: Passthrough from providers on paid plans
Context / input limits: Not documented
SDKs: TypeScript (@openrouter/sdk), Python (openrouter), Agent SDK (@openrouter/agent), OpenAI SDK compatible
Streaming: Yes
Batch processing: Not documented
Structured output: Not documented
Tool / function calling: Yes
Commercial use: See the provider terms
Data retention & privacy: See the provider privacy terms

A free OpenRouter account is not the same thing as free API access. The API allowance above is the provider's documented developer quota.

Verified Oct 19, 2026

Advantages and disadvantages

Advantages

Huge model catalogue
Provider list prices passed through
Free tier for testing

Disadvantages

5.5% (Standard) or 8% (Business) fee on credit purchases
Free model caps are low
Another intermediary in the data path

Privacy: Data-policy-based routing is available on paid plans; requests pass through OpenRouter to the chosen provider.

Real-world examples

A hobbyist testing free models

Sign up and create an API key
Choose a model marked free
Stay within 50 requests per day (1,000 after buying $10 of credits)

free plan works

A developer switching from one vendor

Change the OpenAI SDK base URL to OpenRouter
Swap model slugs to compare outputs
Keep the best model in production

needs API key

A company keeping its own contracts

Add provider keys with BYOK
Route through OpenRouter for fallbacks and analytics
Pay the provider directly

needs API key

OpenRouter alternatives

Together AI

Cloud platform for running, fine-tuning and training open models on serverless APIs or rented GPUs.

no free plan · offers an API

OpenRouter vs Together AI

GroqCloud

Low-latency inference API for open models running on Groq's LPU hardware.

has a free plan · offers an API

OpenRouter vs GroqCloud

Hugging Face

Hub for sharing open models, datasets and demo apps, with paid compute for inference and hosting.

has a free plan · offers an API

OpenRouter vs Hugging Face

Replicate

Run and deploy open-source and custom AI models through an API, paying per second of hardware or per output.

no free plan · offers an API

OpenRouter vs Replicate

Compare OpenRouter with these side by side

Frequently asked questions

Is OpenRouter free?

There is a free plan with 25+ free models, capped at 50 requests per day and 20 per minute. Buying at least $10 of credits raises the free-model cap to 1,000 per day.

Does OpenRouter add a markup to model prices?

Inference is billed at providers' list prices. OpenRouter charges a platform fee when you buy credits: 5.5% on Standard and 8% on Business.

Can I use my existing OpenAI code?

Yes. Point the OpenAI SDK at https://openrouter.ai/api/v1 and use an OpenRouter key.

What is BYOK?

Bring your own key: the provider bills you directly. On Standard and Business, the first $25,000 of list-price inference per month has no OpenRouter fee, then 5% applies.

How many models are available?

The pricing page lists 500+ models from 80+ providers on paid plans.