# Rayline launches ARC, a tiny model that routes AI agent steps to the right model

> Original title: Rayline has launched ARC, a tiny AI model designed to pick which model should handle each step of a task and how much it should think.

- Source: Testing Catalog
- Published: 2026-11-11T20:28:54.000Z
- Bad Influence: https://badinfluence.news/items/bfskg5qs4u74m01gytby2ane1
- Original: https://x.com/testingcatalog/status/2120614769631650146

## Summary

Rayline has launched ARC, a small AI model that picks which model handles each step of a task and how much it should think. In Rayline's DeepSWE tests, ARC scored 68.7% against 69.9% for Opus 5.5 while cutting cost to 7% per task. Users can send a task to ARC, which assigns easy steps to a lighter model and hard ones to a stronger model.

## Post

Rayline has launched ARC, a tiny AI model designed to pick which model should handle each step of a task and how much it should think.

Users can send a task to ARC and have it automatically assign easy steps to a lighter model and hard ones to a stronger model, for a faster, more cost-efficient experience.

In Rayline's DeepSWE tests, ARC scored 68.7% versus 69.9% for Opus 5.5, while dropping the cost to 7% per task.

[Video](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Famplify_video_thumb%2F2120614735225802752%2Fimg%2Fhl_MYRsDO2gmZuel.jpg&mode=thumb&exp=1794700800&sig=bbec98d44d4d15bd)

## Quoted post @davidvgilmore

> In 2015, I built an AI that detects early cancer.
> 
> Today, we’re introducing ARC by Rayline: the world’s first unified model for routing AI.
> 
> > 97.5% of Opus 5.5 quality
> > Up to 1/10th of the cost
> > Fully open-source
> 
> It picks the right model and the right amount of thinking for every step of your agent’s work.
> 
> Try it here and read our technical report below: 
> 
> http://rayline.ai

https://x.com/davidvgilmore/status/2120583960447029699