Models in a few lines
A high-level interface over your chosen backend.
AI Development
Keras development services on the backend you choose
We have no published Keras build. Keras 3 runs on JAX, TensorFlow or PyTorch, and we say plainly that our Python back ends do not train models.
0Published builds on Keras
3Backends Keras 3 supports
3.xVersion the Keras guide pins with keras~=3.0
1Backend you install separately
What we build
What we would take on with Keras
A backend decision
JAX, TensorFlow or PyTorch, picked with your team.
Serving behind an API
A FastAPI service in front of a trained model.
Upgrades to Keras 3
Projects moved off older Keras releases.
Architecture
Where Keras would sit
A Keras model in the slot our AI product fills
Clients
React 19 appAccount briefs and grades
API
FastAPIPython 3.13
Model
Keras modelWhere it would sit
Claude-compatible APISales OS, today
Illustration. A dashed outline marks where the technology would sit. We have no published build on it.
Backends
Keras 3 runs on three backends
The backends Keras 3 supports
JAXAvailable as the jax backend.
TensorFlowAvailable as the tensorflow backend. See our TensorFlow page.
PyTorchAvailable as the torch backend. See our PyTorch page.
Source: Keras, Getting started, keras.io/getting_started, checked 5 November 2026. You also need to install the backend framework.
Builds
The closest builds we can show you
AI product, in-house
Sales OS
Python 3.13 and FastAPI with an AI engine that writes a graded brief through a Claude-compatible API.
1 creditOpens all six tabs of an account brief
Logistics, fleet operations
TruckFleet Labs
A Python 3.13 back end on FastAPI with telematics ingest and PostGIS.
38Drivers on the app by week three
Fit check
When Keras is the right call
Pick Keras when
- You want to build a deep learning model quickly
- You want to keep the backend open
- Your team already knows Keras
Look elsewhere when
- You need a published Keras build to inspect, which we do not have
- A model API already answers the question, as in Sales OS
- Your data is a plain table, where scikit-learn is lighter
Explore
Same job, other tools
Pairs with
Where we used it
Stack
What Keras usually sits beside
TensorFlow
PyTorchPython
Docker
scikit-learn
OpenCVFAQs
What buyers ask first
Which backends does Keras 3 support?
JAX, TensorFlow and PyTorch. You pick one and install it separately.
Do you have a published Keras build?
No. We say so on the first call and show the nearest work, which is a Python 3.13 back end and an AI product that calls a model API.
Keras or PyTorch?
Keras 3 can run on PyTorch, so the question is how much interface you want. We say which fits on the first call, and we have no published build on either.
Who owns the code and the cloud accounts?
You do, from the first commit. Repositories and cloud accounts stay in your name.
Tell us what your model has to learn
Thirty minutes on your data, your backend and your Keras version. No sales pitch.