A Faster API Better Models and Smarter Auto Routing
We improved the ShuttleAI API for better speed and stability, removed unused models, introduced a smarter shuttleai/auto router, refreshed the UI, and expanded context support to 128K tokens.

A Cleaner Model Lineup
We removed most unused and legacy models to reduce confusion and improve reliability.
Less clutter.
Better performance.
More predictable results.
Now it is easier to choose what you need, or just let Auto handle it.
Introducing shuttleai/auto
We added a new intelligent route:
model: "shuttleai/auto"
ShuttleAI Auto automatically selects the best model for your request. You do not need to compare models or guess which one fits.
It looks at:
Task complexity
Length
Domain
Required capabilities
Then routes to the most efficient and capable model available on your plan.
Why use it?
Most cost efficient option
No model guessing
Automatically improves as we add new models
Supports large context requests
If you want the simplest experience, just use:
shuttleai/auto
Read more here:
https://docs.shuttleai.com/models/shuttleai-auto
API Rebuilt: Python → TypeScript
We completely rebuilt the backend from Python to TypeScript.
This means:
Lower latency
Better concurrency
Smoother streaming
More stable under heavy load
Requests should feel noticeably faster and more responsive.
Larger Context Window
We improved context scaling to support up to 128K tokens.
This allows:
Longer documents
Bigger conversations
More memory per request
Better agent workflows
Large prompts now run more reliably without breaking.
Fully Updated Documentation
Docs have been rewritten and reorganized.
Cleaner layout
Better examples
More consistent structure
Easier onboarding
Everything is more clear and easier to follow.
Refreshed UI
We redesigned large parts of the platform:
Updated dashboard look
Cleaner layout
Better responsiveness
Improved navigation
The Playground is also refreshed:
👉 https://shuttleai.com/playground
More UI improvements are rolling out across the site.
Overall Improvements
Faster API
Cleaner models
Smarter routing
Bigger context
Refreshed UI
Better scaling
ShuttleAI should now feel smoother, faster, and easier to use across the board.
If you are already building with us, you do not need to change anything. Just switch to shuttleai/auto and you are good.
More updates coming soon.
