OpenRouter
The unified interface for every model
Last reviewed Apr 4, 2026
OpenRouter provides a unified API for accessing hundreds of AI models from multiple providers. It aggregates models from OpenAI, Anthropic, Google, Meta, Mistral, and many others, offering a single endpoint with automatic fallback routing and provider optimization.
LLM API Pricing
Pay-per-token pricing. Prices shown per 1M tokens.
Prices last updated: October 3, 2026
| Model | Input/1M | Output/1M | |||
|---|---|---|---|---|---|
| $0.017 | $0.112 | ||||
| $0.018 | $0.090 | ||||
| $0.019 | $0.030 | ||||
| $0.021 | $0.062 | ||||
| $0.021 | $0.063 | ||||
| $0.027 | $0.201 | ||||
| $0.028 | $0.056 | ||||
| $0.030 | $0.130 | ||||
| $0.035 | $0.140 | ||||
| $0.037 | $0.170 | ||||
OpenRouter pricing by model
Input, output, and batch rates, plus alternatives, for one model at a time.
Pros & Cons
Advantages
- Largest selection of models through a single API
- Transparent per-token pricing with per-provider comparison for each model
- Automatic failover across multiple providers
- No vendor lock-in — switch models with a single parameter change
- Supports prompt caching for cost savings
- Free plan with 25+ free models (rate limited)
Limitations
- Platform fee on pay-as-you-go usage (5.5% Standard, 8% Business)
- Latency may be slightly higher due to routing layer
- No fine-tuning or custom model training
- In-region routing, SSO/SAML, and contractual SLAs require Business or Enterprise plans
Key Features
Unified Model Access
Single API endpoint for 500+ models from 80+ providers across text, image, video, and audio
Automatic Fallback Routing
Requests automatically route to the best available provider for each model
OpenAI-Compatible API
Drop-in replacement using the OpenAI SDK format
Provider Optimization
Routes to the fastest or cheapest provider based on your preference
Usage-Based Pricing
Pay per token with no minimum commitments or subscriptions
Prompt Caching
Reduced pricing for cached input tokens on supported models
Data Policy-Based Routing
Model and provider policies ensure prompts only go to the models and providers an organization trusts
In-Region Routing
EU and US in-region routing on Business and Enterprise plans
Bring Your Own Key (BYOK)
Use existing provider API keys through OpenRouter, with a monthly fee-free allowance before a 5% fee applies
Pricing Options
| Option | Details |
|---|---|
| Free | 25+ free models from a small set of providers, rate limited, with community support |
| Standard (pay-as-you-go) | Per-token model pricing plus a 5.5% platform fee, with no subscription |
| Business (pay-as-you-go) | 8% platform fee; adds EU/US in-region routing, workload identity federation, and higher workspace limits |
| Enterprise | Custom agreements with fee discounts, SSO/SAML, contractual SLAs, invoicing options, and a dedicated support team |
| Prompt caching | Discounted rates for cached input tokens on supported models (paid plans) |
Availability & Support
Regions
Global — routes to providers across multiple regions for optimal latency
Support
Community support on the Free plan, email support on Standard and Business, dedicated support team on Enterprise; documentation and Discord community
Getting Started
- 1
Create an account
Sign up at openrouter.ai; an organization for your team can be set up later
- 2
Buy credits
Purchase credits usable with any model or provider (credit card, crypto, and more)
- 3
Get API key
Create an API key from your dashboard
- 4
Make API calls
Use OpenAI-compatible endpoints or the OpenRouter Client and Agent SDKs with your key
Frequently Asked Questions
Which models does OpenRouter serve?
Check the pricing table above for the models OpenRouter currently serves and their per-token rates.
How do I get started with OpenRouter?
Create an account, Buy credits, Get API key, Make API calls
What are OpenRouter's main advantages?
OpenRouter's main advantages include: Largest selection of models through a single API, Transparent per-token pricing with per-provider comparison for each model, Automatic failover across multiple providers, No vendor lock-in — switch models with a single parameter change, Supports prompt caching for cost savings, Free plan with 25+ free models (rate limited).
What are OpenRouter's limitations?
OpenRouter's main limitations include: Platform fee on pay-as-you-go usage (5.5% Standard, 8% Business), Latency may be slightly higher due to routing layer, No fine-tuning or custom model training, In-region routing, SSO/SAML, and contractual SLAs require Business or Enterprise plans.
Compare Providers
Find the best prices for the same GPUs and models from other providers