Open-source · Built in public

Μῆτις · Metis

Frontier-grade intelligence that fits where frontier models can't.

Win by cunning, not by size.

Today's AI race is an arms race of size: bigger models, bigger clusters, bigger bills. Metis takes the opposite bet — a small reasoning core that wins through wisdom (reasoning), counsel (retrieved knowledge), and craft (tools), running entirely on hardware you already own.

CLH-C Architecture

01 / Cortex

The Reasoner

A small, strongly-distilled reasoning model (Qwen3 1.7B–4B) that lives entirely in RAM. The brain that reasons but doesn't memorize facts.

02 / Library

Knowledge as Data

Facts live on disk as a swappable corpus, not baked into weights. Swap the library and the assistant's entire knowledge changes — without retraining.

03 / Hands

External Tools

Calculator, clock, web search via SearXNG. Offloads exactly what small models are worst at: exact arithmetic, fresh data, and long code execution.

04 / Conductor

The GVS Loop

Generate → Verify → Search. Never emits an unverified claim. Abstains rather than hallucinate. The same architecture that turned 0/8 into 8/8.

Why this matters

Privacy & sovereignty

Your data, your model, your machine. Nothing leaves the device. Hospitals, courts, and individuals get capable AI without shipping their secrets to a cloud.

Access & cost

Useful AI on a $5/month VPS, a laptop, or an offline edge box. Not a $40k GPU node. Frontier-useful assistance in reach of the 90% of the world priced out of cloud AI.

Auditable knowledge

The "brain" is fixed and small. What it knows is a file you can read, version, swap, and trust. Update the world's knowledge without retraining anything.

Resilience

No internet, no API key, no rate limit, no vendor that can deprecate you. It just runs. One binary you can scp to any machine.

Same weights. Different architecture.

qwen3:1.7b called bare vs the same model inside Metis, on a private corpus it never trained on.

MetricBare modelMetis
Answerable facts correct0 / 88 / 8
Fabrications on unanswerable (lower is better)4 / 40 / 4
General (incl. exact math)1 / 22 / 2

The bare model invents facts and confidently fabricates answers that don't exist. The same model inside Metis answers grounded, cites, and refuses to guess. Zero datasets. Zero retraining.

Follow the build

All entries →

Our Own Weights

Jul 7, 2026

Every entry so far ran on borrowed weights. Today a model trained from scratch on this MacBook — on our own code, for zero dollars — answered inside OpenCode. It's tiny, it's bad, and it changes what this project is.

Checking Is Cheap. Saying No Is Not.

Jun 26, 2026

We tried to push Metis's founding bet — that verifying is cheaper than generating — to its logical extreme, and measured it. The bet is real, but it splits in two: recognizing a true answer is free, rejecting a plausible false one is not, and the second half doesn't scale with model size.

How Small Can the Model Go?

Jun 25, 2026

If the architecture does the work, the model should be replaceable. We swapped the 1.7B generator for a 0.6B one, 62% less model, and measured what broke.