Edge Model Foundry

Your next edge product does not need a bigger processor.

It needs several small ones, each carrying a different skill. That is how speech, vision and condition monitoring get into a device whose compute costs four dollars a part, and how you find out it fits before you commit to hardware.

Start free See it running

hear-and-say. Audio in, a board decides, speech out. Recorded from the running editor.
21models, measured
on real silicon
3.24×from adding
three parts
$4per part, at
volume
0data leaving
the device
OpenApache 2.0 runtime,
yours to keep

Why teams choose this

What changes when intelligence is something you add, not something you specify.

Before you commit to hardware

Find out it fits while the design is still cheap to change

The expensive moment in an edge programme is discovering, after the board is laid out, that the model does not fit the part. Send us the model and get back the flash it occupies, the working memory it needs and the milliseconds it takes, measured on the silicon you are considering.

That answer currently costs an engineer a week and a development board. It should cost an afternoon.

Every model we publish already carries those three numbers
Without a hardware revision

Add a skill by adding a part

Halfway through a programme the requirement changes: it needs to hear as well as see. On a single processor that is a re-spin, a new thermal budget and a schedule slip.

Here it is another inexpensive part carrying another model, wired into the same graph. Capability scales by addition, and the parts you are not using draw almost nothing.

Hearing, language, speech, vibration and electrical sensing together exceed what any one part in this class can hold
Compliance as the entry condition

Nothing leaves the device, so the conversation is shorter

Plants that will not put a microphone on a network. Defence programmes that will not accept a call home. Products where a recording reaching a cloud is the objection that ends the sale.

Inference happens on the part. There is no gateway, no network dependency and nothing to exfiltrate, which turns a procurement blocker into a line in the datasheet.

Air-gapped operation is the default, not a deployment option
A different business case

Instrument forty points for what four used to cost

Condition monitoring is well understood and priced as a capital project, because the sensing hardware is expensive. Change the cost of a sensing point and the arithmetic changes with it.

Covering a whole line stops being a proposal that needs board approval and becomes an operating line item, which is usually the difference between a pilot and a rollout.

Learned-normal monitoring needs no labelled fault data to start

The idea

More parts, not bigger ones.

Splitting work across cheap parts to finish sooner is ordinary parallelism, and a larger processor would do the same. That is not the argument.

The argument is that capability composes. Each part holds one skill. Together they do what none of them could hold, and no single processor in this class runs the whole system at any clock speed.

This is an old idea that was priced out of reach. Minsky argued that intelligence is what a large number of small, individually unintelligent processes produce between them, and Hillis built the machine that took the idea seriously. It cost what a research instrument costs, because the processors were expensive. They are not any more.

How machines compose

3.24×
More coverage

A camera frame split into twelve overlapping windows across four parts, each answer returned to its origin.

0.838
Better decisions

Two presence models pooled, against 0.769 for the stronger one alone. The cost was one more part.

4×
Faster rollout

Writing a model to five parts runs at 444 KB/s against 108 KB/s to one. Fleets update in parallel.

1,604 B
Smaller than you expect

A complete prompt-to-image generator, with no dynamic allocation anywhere in the path.

Try it on five parts, free, for as long as you like.

The free tier is five boards rather than one, because a single part cannot demonstrate the thing this is for. The whole argument is what happens when several narrow models are wired together, and a trial that cannot show that is not a trial.

The runtime stays Apache 2.0 whatever you decide afterwards. Prove a machine here and run the identical graph air-gapped on your own hardware.

Free
Tessera Core

The whole runtime, unlimited boards, commercial use, no expiry and no call home.

$249/mo
Bench to pilot

A hosted place to build machines, with fit data measured against your parts.

See what is included