Samsung's Tiny Model Made Scale Look Wasteful

The AI industry keeps returning to the same lever: bigger models, more parameters, larger training sets. Samsung’s 7-million-parameter model pulled a different one. On specific abstract-reasoning benchmarks, it beat models with thousands of times more parameters.

The Tiny Recursive Model (TRM) doesn’t solve problems in one pass. It loops: draft an answer, check the logic, rewrite, repeat up to 16 times. An internal “scratchpad” critiques its reasoning, catching mistakes before they compound.

On the ARC-AGI benchmark, a set of abstract reasoning tests that trip up even the best LLMs, TRM scored 45% on ARC-AGI-1 and 8% on ARC-AGI-2. Better than DeepSeek-R1, o3-mini, and Gemini 2.5 Pro, all running billions of parameters. On Sudoku-Extreme with 1,000 training examples, it hit 87.4% accuracy versus its predecessor’s 55%.

Alexia Jolicoeur-Martineau, the Samsung researcher behind TRM: “The notion that one must depend on extensive foundational models trained for millions of dollars by major corporations to tackle difficult tasks is misleading.”

You don’t write down the first answer that pops into your head when solving a hard problem. You draft something, spot the holes, rethink it, try again. TRM does this in a two-layer neural network. Traditional LLMs generate answers in a single forward pass. An early mistake spreads through the entire response.

TRM won’t write your emails or debug your code. It handles structured, grid-based reasoning: Sudoku, mazes, abstract pattern recognition. For these problems, the paper shows that a 7-million-parameter model can beat much larger cloud models. That is a narrower and more interesting result than claiming small models win at everything.

I’ve been testing Apple’s Foundation Models framework. What I like most is the privacy. It runs entirely offline. Your prompts stay on your device. Apple doesn’t use your data to train models.

Cloud-based LLMs log every query you send. Apple’s 3-billion parameter model runs natively in iOS 26, iPadOS 26, and macOS 26. It feels like it’s mine, not something I’m renting access to.

Samsung’s TRM is open source under an MIT license. Apple’s Foundation Models framework also gives me a practical way to build with an on-device model. Neither replaces the general-purpose models I use for coding, but both make the same point: for a bounded problem, I would rather run the smallest model that can do the work.