One API that automatically picks and runs the best open-source AI model for every prompt.
Start GeneratingSubmit your AI request
Analyzes and selects best-fit model
Generates output with chosen model
Returns your AI response
Routes to powerful models when needed for complex tasks
Cost-effective models used when possible, saving you money
Built on the latest and best open-source large language models
Works with all AI apps that support OpenAI compatible custom APIs
Clear documentation and simple examples to get started quickly
Earn recurring income when your app's users subscribe to our API
Pick a plan to suit your usage.
/ month
No hidden charges, cancel anytime.
/ month
No hidden charges, cancel anytime.
/ month
No hidden charges, cancel anytime.
/ month
No hidden charges, cancel anytime.
/ month
No hidden charges, cancel anytime.
/ month
No hidden charges, cancel anytime.
Note: Priced in USD and tax exclusive.
Credits are consumed per API request. Each request uses 1-5 credits depending on computational requirements and which model is used - simple, short responses use 1 credit while complex, longer responses can use up to 5 credits. Credits refresh monthly and unused credits don't roll over.
If you run out of credits, you have three options: upgrade to a larger plan (we'll prorate the cost), purchase additional top-up credits, or wait for your monthly credit refresh. Your account remains active either way - you'll just receive an error response for new requests until you have credits available.
Yes, if you're using an AI app that supports custom API providers, you can use our service. Many popular AI apps allow you to plug in alternative AI providers - check your app's settings or documentation for more information.
You can start monetizing your AI app with our API. We handle all the complex parts including model selection, billing, and infrastructure while you earn recurring revenue from every user. Our OpenAI-compatible endpoint makes integration simple, and our easy to follow documentation gets you up and running quickly. It's the easiest way to turn your AI app into a revenue-generating business.
Our router analyzes each prompt and automatically selects the most appropriate model. Complex tasks get routed to powerful models, while simpler ones use efficient models. This optimizes both performance and cost - you only use powerful, more expensive models when needed.
While our AI router automatically selects the best model for each prompt, you can set a routing preference with each request: Balanced (default), Powerful, or Efficient. Powerful mode favors more powerful models for best results but uses more credits, while Efficient mode prioritizes lighter models when possible to minimize credit usage.
You can cancel your subscription anytime from your account dashboard. Your service will continue until the end of your current billing period, and you won't be charged again.
After payment, you'll be redirected to your account dashboard. From there, you can grab your API key and either connect it with any supported app or head over to the documentation to start generating.
@meta-llama
Balanced
@deepseek-ai
Powerful
@meta-llama
Powerful
@qwen
Balanced
@mistralai
Efficient
@mistralai
Efficient
@nvidia
Balanced
@deepseek-ai
Powerful
@deepseek-ai
Efficient
The collection above is for informational purposes only. Refer to the documentation to get an accurate list of models used in our generative AI API.
Integrate our API into your AI application and earn recurring revenue from every user.
Apply NowStay up-to-date with the latest developments from us