Skip to main content

Dev Station Technology

AI Development

ONNX development services for models that must run on a phone

We have no published ONNX build. Our driver app is native iOS and works with no signal, and ONNX Runtime also runs on iOS, which is where a model would sit.

0Published builds on ONNX
1.23.2Marked latest on the ONNX releases page
1.23.0Released 18 September
6Platforms ONNX Runtime lists
What we build

What we would take on with ONNX

Models on the device

Inference on a phone that works with no signal.

Models behind an API

A FastAPI service that loads an ONNX model.

Model export

A trained model exported from your framework to ONNX.

Tests on real devices

The same model checked on each platform it ships to.

Architecture

Where ONNX would sit

A model on the phone beside the sync queue
Phone
iOS logo
Native iOS appSQLite on the device
ONNX logo
ONNX modelWhere it would sit
API
Python logo
FastAPIPython 3.13
Data
PostgreSQL and PostGISStops and geofences
RedisSync queue

Illustration. A dashed outline marks where the technology would sit. We have no published build on it.

Releases

Recent ONNX releases

The latest ONNX releases
15 June

ONNX 1.22.0

A minor release.

18 September

ONNX 1.23.0

The next minor release.

29 September

ONNX 1.23.1

A patch release.

6 October

ONNX 1.23.2

Marked latest.

Source: ONNX releases on GitHub, checked 6 November 2026. The page shows day and month without a year. ONNX Runtime describes itself as a production-grade AI engine for training and inferencing.

Builds

The closest builds we can show you

TruckFleet Labs dispatch console and driver app built by Dev Station
Logistics, fleet operations

TruckFleet Labs

A native iOS driver app that works with no signal, with a FastAPI back end behind it.

71Queued records delivered after an outage
Sales OS account research product built by Dev Station
AI product, in-house

Sales OS

A Python 3.13 and FastAPI product whose AI step calls a Claude-compatible API.

1 creditOpens all six tabs of an account brief
Fit check

When ONNX is the right call

Pick ONNX when

  • A model must run on a phone or in a browser
  • You trained in one framework and deploy in another
  • Predictions must work with no signal

Look elsewhere when

  • You need a published ONNX build to inspect, which we do not have
  • A model API already answers the question, as in Sales OS
  • You have no trained model yet
Stack

What ONNX usually sits beside

Python logoPython
PyTorch logoPyTorch
TensorFlow logoTensorFlow
iOS logoiOS
Android logoAndroid
Docker logoDocker
FAQs

What buyers ask first

Which ONNX version is current?

The ONNX releases page marks version 1.23.2 as latest, published on 6 October. Versions 1.23.1 and 1.23.0 came out on 29 September and 18 September.

Do you have a published ONNX build?

No. We say so on the first call and show the nearest work, which is a native iOS driver app and Python back ends.

Where does ONNX Runtime run?

Its site lists Linux, Windows, Mac, iOS, Android and web browsers.

Who owns the code and the cloud accounts?

You do, from the first commit. Repositories and cloud accounts stay in your name.

Tell us where your model has to run

Thirty minutes on your model, your devices and your signal. No sales pitch.

Let's Talk