What it is
A living database of common tasks that says — with evidence — whether a leading local LLM (like Gemma) can safely do each one, which local models clear it, and what to use instead when none can.
DoesItLocal
Can a local LLM safely do this task — and if not, what should you use instead? A living, evidence-backed answer per task.
Read the introduction See the roadmap
How it works
Each task carries a verdict (safe for local / local with a check / needs a bigger model / needs more data), backed by eval runs and manipulation-resistant practitioner votes — never a bare leaderboard rank.
Start here
Section titled “Start here” Introduction What it is and the problem it solves.
Who it's for Developers routing work between local and frontier models.
How it works From a task to a verdict, end to end.
Roadmap What ships first, and in what order.