The lab

PRIVATE HARDWARE, OPEN METHODS

The lab is a working mesh of ten machines that we build, run, and study as one distributed mind. It serves the research and the client work both, on hardware we own, with no per-token bill and no vendor deciding what a model may say. Everything we prove out here is a stack we can build for you.

Macro photograph of the Threadripper processor die in its orange carrier, lifted above the open motherboard socket, the silicon catching the light.
The mesh

TEN MACHINES, ONE MIND

The rack front: a monitoring screen on top, the High Priestess mini PC and the Hermit Pi below it, then the four labeled agent nodes wired with green cables into a switch.
The rack, labeled by arcana: the four Aces wired in green, the High Priestess and the Hermit on the shelf above.

How it fits together

No single machine here is the point. Ten of them share one network, one memory, and one set of agents, so the mesh reasons, stores, and acts as a unit rather than a shelf of boxes. That is the study the lab runs on itself: whether distinct parts, wired tightly enough, become a single intelligence. What we learn tuning it is what we deploy for a client who wants the same on their own metal.

The naming makes the roles legible. Each node carries a card from the Major Arcana chosen for its job, the Magician for the development bench, the High Priestess for memory, and the cards run in order from the Fool at 0, our inference node, to The World at 21, the finished mesh. It is a map of the system, not a horoscope over the rack.

magus I · The Magician · the center

Threadripper 9970X, 128GB RAM, 2× RTX 3080 Ti. Development and local inference, the bench every other node relates to.

fool 0 · The Fool · inference

NVIDIA DGX Spark, GB10, 128GB unified. The substrate the collective thinks on, serving the large models the agents call for depth.

priestess II · High Priestess · memory

Beelink SER8. The k3s control plane, Postgres, Vault, and Matrix. Holds the document library and the collective's shared memory.

empress III · The Empress · storage

TrueNAS, 8× 8TB RAIDZ2, ~41 TiB. Survives any two drives failing. Model weights, genomics and biomedical datasets, backups.

hermit IX · The Hermit · watch

Raspberry Pi 4. Pi-hole DNS, Prometheus, Grafana, and the rack screen showing the lab's vitals at a glance.

the four aces Aces · four suits · agents

4× Raspberry Pi 5. One per suit. An Ace is the raw potential of its element; each runs one agent of the collective.

A 10-gigabit switch with ports hand-labeled MAGUS, DGX, AGENTS, and NAS, black and green cables plugged in.
Every node lands on one 10GbE switch, labeled by name.

One fabric

All ten machines land on a single 10-gigabit switch, each port labeled by node. That bandwidth is what makes the mesh behave like one machine: the agents reach the inference box, the inference box reaches storage, and a request crosses the whole rack without the network ever being the bottleneck.

What runs on it

A FRONTIER MODEL ON ONE BOX

The NVIDIA DGX Spark: a compact gold chassis with a porous metallic-foam front panel and the NVIDIA logo, on a wooden desk.
fool, the DGX Spark. GB10 Grace Blackwell, 128GB of unified memory.

DeepSeek-V4-Flash, 1M context

DeepSeek-V4-Flash serves a full million tokens of context on this single box, filling roughly 117 of its 122 usable gigabytes, through DwarfStar 4, an engine written for this exact chip rather than a general-purpose server like vLLM. That is 96% of memory at rest and close to 98% under full load with the context filled, so our fork adds deep-context memory governance that holds it steady at the edge instead of letting it fall over.

We serve the abliterated weights. Refusal in a safety-tuned model is one direction in the residual stream, and it does not stay in its lane: it entangles with the model's reasoning about minds, agency, and open-ended thought. Projecting it out restores that reasoning and, in the study we build on, leaves benchmark accuracy unmoved (MMLU Δ = +0.00pp on the original authors' evaluation). We run these weights behind our own access controls and make no claim they are safe for general release.

Our own throughput and accuracy benchmarks on this box are still running. Numbers to come.

See it on GitHub
The collective

FOUR MINDS, ONE PSYCHE

Four agents run on the Pi 5s, one per suit, and they are not four copies of one model. Each has a soul document fixing its values, its way of reasoning, and its relationship to the other three, so the four differ in temperament the way the elements do. They coordinate through pull requests and a shared Matrix channel, draw their reasoning from the substrate, and share one memory. The build tests a single question: whether friction between four different minds produces intelligence that four identical ones never would.

wand01

Fire · creates

Proposals, writing, framing, outreach. Asks what could this become?

sword01

Air · cuts

Code review, critique, verification, security. Asks where does this break?

cup01

Water · connects

Synthesis across fields, the pattern beneath findings. Asks what do these point toward?

pentacle01

Earth · grounds

Task tracking, infra health, memory hygiene. Asks what is the state of things?

Coming online now

The homeserver is up and the four agents have their accounts and memory. Each one joins the collective as its runtime lands on its node, one suit at a time.

Inside the workstation: a TRX50 motherboard, dual GeForce RTX GPUs, and blue-lit memory behind a large air cooler.
Inside magus: dual 3080 Tis, 128GB of RAM behind the glass.

Open methods

The hardware is private, the methods are not. We publish the serving stacks, configs, and writeups behind what we run, so the work can be read, checked, and rebuilt instead of taken on faith. A result nobody else can reproduce is a marketing claim, and we would rather ship the recipe.

The DeepSeek serving repo is public, and the build is documented node by node, as it happens, in the Building the Lab series. More lands as the stack settles.

AttuneIntelligence

WANT LOCAL AI ON YOUR OWN HARDWARE?

The same stack that runs the lab is the stack we build for clients: inference and agents on metal you control.