Chat with any AI model.
Pay only what you use.

Bring your own API key. No model limits. No daily usage. Just a clean interface for the models you want to use.

Pick your model

Switch between GPT-5.2, Claude, Gemini, Llama, and more. See capabilities at a glance.

Claude Opus 4.5
anthropic/claude-opus-4.5
GPT-5.2
openai/gpt-5.2
GLM 4.7 Flash
z-ai/glm-4.7-flash
claude-opus-4.5•42.8 tok/s•847 tokens•$0.032688
Model:$0.016688
Search:$0.015000
Extract:$0.001000
Total:$0.032688

See what you spend

Real-time cost tracking per message. Know what you're paying for tokens, search, and file processing.

Your keys, your choice

Store API keys locally in your browser, or encrypted on our servers. You're always in control.

OpenRouter API KeyConfigured
sk-or-v1-••••••••••••••••
Browser Only
Stored locally
Encrypted Server
Selected
read up on how AI works for me
Sources used
Explained: Neural networks | MIT NewsWhat Is a Neural Network? | IBM
Extracted pages
Explained: Neural networks | MIT News
This article from MIT News explains that...

Search smarter

Powered by Parallel.ai — the highest accuracy AI web search. Search and extract content in parallel for faster answers.

Frequently Asked Questions

Pricing

Free

Everything you need to get started.

$0
  • 100 threads
  • 10 MB storage
Recommended

Plus

For power users who need more.

$2.99/mo
  • Unlimited threads
  • 100 MB storage
  • Priority support

Both plans include encrypted messages and Bring Your Own Key support. You always pay for your own API usage through OpenRouter.