Amazon Web Services has released an open-source decision model inspired by TypeSafe's Jev, because nothing says innovation quite like watching someone else's homework and deciding to do your own version of it. The Strands Decider 2B, released the same week OpenAI announced a suspiciously similar offering, promises to be a high-speed, low-cost way to sort between pre-decided options - essentially a multiple-choice machine that tells you how confident it is in its answer. It's fully open-sourced, available now, and small enough to run locally, which is tech-speak for "you don't need a data center to use it."

The project began when Amazon distinguished engineer Marc Brooker saw Jev and thought, "I could build that." Turns out he could - the homebrew version briefly hit the top spot on the Jevbench ranking for models of its size, which is like winning a participation trophy that actually means something. Amazon engineers then cleaned it up and released it through Strands Labs, the company's organization for developing new AI agent deployment tools.

Brooker says the need emerged from conversations with AWS customers whose agentic workflows didn't always require the capability or cost of a fully-featured LLM. "What originally piqued my interest in this class of models was that they make a perfect decider for a workflow step - 'what is the next thing for me to do here, based on where I am?'" he told TechCrunch. He added that it offers customers a workflow step that's more reliable thanks to confidence scores and a closed domain of answers, with lower latency and potentially lower cost. In other words: sometimes you just need a traffic cop, not a philosopher.

Like other decision models, Strands Decider is built on the "torso" of an LLM - specifically Qen3.5-2B - but instead of generating text, it delivers calibrated choices. TypeSafe named their model Jev after economist William Stanley Jevons, invoking his theory that falling costs can increase demand. Because nothing says "AI revolution" like naming your product after a 19th-century economist.

The fact that dozens of similar models have popped up since TypeSafe debuted its idea shows wide interest, but also raises the question of how valuable they can be. Brooker suggests the challenge is optimizing speedy decision-making without compromising intelligence. "There is a very careful balance to be found where you want to push its performance on accuracy and calibration on these kinds of tasks, without degrading its performance on understanding different languages, on having the kind of knowledge it has, which is what makes it general purpose and interesting and useful," he said. So, the goal is to make a model that's both fast and smart - a combination that has historically eluded both AI and humans alike.

Still, Brooker doesn't expect frontier labs to dominate the space, especially since the cost to build something interesting is only in the hundreds or thousands of dollars. For their part, TypeSafe executives say they're keeping their heads down and improving future models. "I get that people think it's a gold rush, but they might be underestimating the difficulty of making the models actually smart," CEO and founder Diogo Almeida told TechCrunch, adding that he didn't see real competition emerging yet. "The current batch seems more like ML people wanting to implement a cool architecture than a team deeply dedicated to making intelligence useful." Which is a polite way of saying: nice try, but call us when you've actually built something.