The fastest way to build ML-powered applications.

Avoid deployment nightmares and hacking together unscalable APIs and front-ends. Create production-grade apps with ML in minutes and start iterating with users today.

🔮 By joining, you’ll get free usage in exchange for your insights.

python

Extract content from Wikipedia article

def pre_process(api_input): url = api_input['wikipedia_url'] content = get_content(url) return {'model_input': content}

Run named-entity recognition model

* Model running.....

Translate output to enriched text

def post_process(out): predictions = out['predictions'] store_predictions(predictions) return {'display': enrich(predictions)}

https://en.wikipedia.org/wiki/Barack_Obama
{ "model_input": [ "Barack", "Hussein", "Obama", "II", "is", "an", "American", "politician", ...] }
1.
2.
 
 
 
3.
{ "predictions": [ { term: "Barack", type: "name" }, { term: "Hussein", type: "name" }, ... ] }
1.
2.
3.
...
98.
99.
Barack Hussein Obama II American 44th United States
Heading
Text display
Text area
Button
NER output
Button
NER output
NER model
Input
Barack Hussein Obama II is an American politician and attorney who served as the 44th president of the United States from 2009 to 2017. A member of the Democratic Party, Obama was the first African-American president of the United States. He previously served as a U.S. senator from Illinois from 2005 to 2008 and as an Illinois state senator from 1997 to 2004.
Analyze text Analyzing... Analyzed!
Output
Barack Hussein Obama II American 44th United States 2009 2017 Democratic Party Obama African-American United States Illinois
Start with a model

Deploy your existing TensorFlow, PyTorch or scikit-learn models with BaseTen's client API or choose from BaseTen’s library of pretrained models to get you started.

Build your APIs

Using Python and a host of third-party integrations, instantly deploy endpoints to power your application.

Craft UI without JavaScript or HTML

Drag-and-drop pre-defined components to create a custom UI for your application, all without learning React.

  • Build APIs
  • Drag and drop view builder
For data scientists
Build scalable, user-centric applications

There's no need to get bogged down in model deployment, scaling APIs, and building user interfaces. BaseTen allows you to start embedding machine learning into real, shareable apps in minutes; be it a simple prototype to showcase the power of your model or a model-driven workflow for content moderation.

For engineers
Jumpstart ML efforts with state-of-the-art models

BaseTen provides pretrained models for common use cases. Create APIs for classifying unseen data or analyzing sentiment without re-inventing the wheel. Integrate your existing data sources, write pre and post-processing code within our serverless infastructure, and even chain the outputs of several deployed models.

  • Model zoo
  • UIs for model zoos
models
Pre-trained models for images, text, and audio come ready to use
pytorch tensorflow
Support for PyTorch, TensorFlow, scikit-learn models and more
In build Postgres database
Access to built-in Postgres database
Infrastructure
No infrastructure configuration required
kafka mysql postgres redis sowflake
Bring data in from other databases and systems of record
sharing
Apps can be shared with individuals or made public
— Trusted by data scientists and engineers at —
alex amir mike paige pankaj phil sam suren tuhin

The BaseTen team

We're a small group of engineers and designers figuring out how to make our dent with thoughtfully designed software. We’re an internationally and intellectually diverse group of passionate builders who try not to take ourselves too seriously. We’re backed by top-tier investors and prominent angels and looking to grow our team. Join us!