Fuses several models per query. Fans the call out, blends the answers, bills for all of it. An algorithm guessing at consensus.
Open-source · MIT · Semantic routing
Not the best model overall. The best model for this task. Debug goes to the debugger, prose to the writer — decided per request, from data.
The open-source objection, up front
Trade a little quality for sovereignty and a zero bill. A good-enough model instead of the frontier one.
That's the assumption CoeOS is built to break.
Not good-enough. Better.
A single closed model is a generalist. Strong everywhere, exceptional nowhere. It drags its weak axes into every request whether the task needs them or not.
CoeOS doesn't run one model. It runs the specialist for each skill — the open-source model that beats every rival on that exact axis, closed models included. Debugging to the one that dominates debugging. Legal analysis to the one that dominates legal. Fifteen disciplines, fifteen champions, one endpoint.
That's the whole thesis.
Earned empirically, not claimed
Not MMLU. Not a suffix on a model card. Not a number to screenshot. Dozens of models run through the same battery, scored discipline by discipline, until a real scoreboard emerges — who actually wins where, with the receipt attached. Every routing decision traces back to a score. Not a hunch. A result.
The TMB benchmarks
Thirty-one substantial tests define fifteen skills. Ten separate tests decide the tool-calling champion alone; behind them, one clean setting the pilot can trust. These aren't trivia. They dissect how a model reasons, what it values, how it holds up under a concrete case any real user might hand it — the kind of task that separates a model that looks capable from one that survives real work.
A pilot, not an aggregator
Fuses several models per query. Fans the call out, blends the answers, bills for all of it. An algorithm guessing at consensus.
Reads the request, understands what it is, and steers it — whole, once — to the one model proven best at it. One classification. One call. The answer returns untouched, in the caller's own format.
Semantic routing, not statistical mush. It knows what the request needs before it picks who answers.
Open-source · MIT