Available Now

Edge-Class Backbone

Linear-time architecture for on-device inference

A sequence model with linear complexity in sequence length — the compute budget that lets 15M parameters run on a smartphone, a feature phone, or a sensor. Pair with Sozisi and the tokenizer and you get GPT-4-class NLU at a fraction of the cost, fully offline.

Properties

What this core unit gives you

The guarantees this core unit provides when composed into a production system.

<50ms inference on commodity hardware

No GPU required for production workloads

Runs on Android, iOS, and embedded Linux

24MB footprint — fits alongside your app

Composes With

The other core units it pairs with

Core units are decoupled by design. These are the ones we've validated in production compositions — but you can compose with your own stack too.

Compose Edge-Class Backbone into your stack.

Available via API, on-prem license, or as part of a composed packaged product. Talk to us about the right entry point.