Encode your knowledge into machine learning models that can automate text-based workflows at scale with human-level quality. Build your own models from scratch, retrain our world-class models for your specific task, or use Primer models off-the-shelf.

Primer Automate


Build custom models quickly to automate tasks

Train your own models to execute your organization’s repetitive text-based tasks, such as classifying text, identifying entities within a document, or summarizing information — all with human-level precision.

Sample Link(Opens in a new window)


Customize our best-in-class models for your specific needs

Use Primer’s wide range of off-the-shelf models as a starting point and retrain them on your own data to perform your specific tasks. Deploy custom models faster with less training time needed.

Sample Link(Opens in a new window)


Go straight from labeling to deployment with an all-in-one, easy-to-use system

Configure, label, train, and deploy your models with the push of a button and get started using AI right away. With Automate, there is no need to move between different platforms; it’s all here in one place.

Sample Link(Opens in a new window)


Integrate your models into your existing systems

Deploy specific algorithms to support your existing systems and data workflows using the Primer API. You can easily generate an API key for every model you create, so you can innovate faster and build your own unique NLP solutions.

Sample Link(Opens in a new window)


Label data quickly and easily

Get the flexibility you need to train your NLP model — from summarization to entity recognition to classification of the relationships between entities. Primer Automate’s intuitive labeling interface makes labeling and other NLP tasks easy.

Sample Link(Opens in a new window)


Understand and monitor the models you create

Check how your models are working with the push of a button. Simple visualizations allow you to interpret the models, see what data they are using to make their decisions, and audit them if they are making mistakes.

Sample Link(Opens in a new window)


Access the world’s leading NLP models in one place

Prototype your NLP workflow with Primer’s pre-trained, domain specific models that work out-of-the-box. Simply log in, upload your data, and choose from dozens of models, or generate an API key and deploy these models on your own data.

Sample Link(Opens in a new window)

Primer Automate


Custom model:
Triaging consumer feedback

Consumers today provide feedback to brands across a plethora of online platforms. Customer service teams use Primer Automate to build a multilabel risk classifier model that helps them process customer feedback in social media, blogs, and other channels. The model constantly monitors for brand and product mentions and will sort feedback into designated groups, such as complaints, improvements, customer churn, or harassment. They can also integrate the model into their social listening, dashboarding, or customer support tool via the Primer API.

Sample Link(Opens in a new window)


Retrained model:
Drug side effect tracking

Pharmaceutical companies need to monitor reports of side effects from a particular medication that are contained within a massive number of clinical patient notes. Pharma analysts use Primer Automate to quickly train their own version of Primer’s cutting-edge Named Entity Recognition (NER) model to identify product-specific terms, such as product name, illness treated, or symptoms. Primer allows analysts to specify their own labels or correct ones to better tailor the model to their unique search intent.

Sample Link(Opens in a new window)


Off-the-shelf model:
Terrorist event identification

With thousands of documents to process every minute, identifying terrorist events manually is an expensive process. National security intelligence teams use Primer’s Text Classification model to help them identify terrorist events within the documents in their data set. With this model built, their analyst teams can deploy it onto their real-time streaming data to identify terrorist events from around the world. This model can also be coupled with the Named Entity Recognition model to understand the organizations involved in the terrorist event.

Sample Link(Opens in a new window)

Primer Engines

Frequently Asked Questions

  • What models can I train?

    Engines are pre-trained, proprietary models built with Primer’s state-of-the-art NLP (Natural Language Processing), NLU (Natural Language Understanding), and NLG (Natural Language Generation) technology. Engines enable humans to automate their reading and writing tasks at scale with human-level precision and discover insights that were previously impossible to find.

  • What models can I use off-the-shelf?

    Primer provides a variety of off the shelf models such as COVID classifiers (as seen on as well as Primer’s best in class Named Entity Recognition. New models are being added continuously.

  • Can I upload my own training data with my own labels?

    Yes, you can customize your models with your own training data which. Just put your labelled data into a spreadsheet or CSV file and upload to Primer Automate – it will take care of the rest.

  • What is the performance of the models you have?

    Models our customers create using Automate will have their own performance metrics available (like F1 Score, Precision, Recall and Accuracy). Off-the-shelf models model performance is provided as well. Most customers can train a model with over 90% accuracy for their use case in a few days.

  • How do I access the models I create?

    Your team’s models live in your account and can be collaborated on by the entire team. When you’re ready to use them you can get predictions on datasets you’ve uploaded via a UI that will visualize your results and return your machine-labeled data to you.

  • Can I access the models via an API?

    Yes – all off-the-shelf and trained custom models come with API access. After you train a custom model, it will be automatically deployed and your API details will be available for you to hand off to your developers to integrate this model in your product or processes.

  • Can I update my model with more training data at a later date?

    Your custom models can be retrained at any time with additional data or changed labels. Models and data can drift with the seasons and as language evolves, so Automate makes it easy to run a new training cycle every so often to keep your model fresh and performant.

  • How does your platform deal with mislabeled data?

    Accurate training data is key to the success of your custom model and Automate helps you create highly accurate labeled data in a variety of ways, chiefly, Automate uses active learning to reduce the amount of labels you need so you can use subject matter experts to provide correct examples. We also offer label peer review so teams can collaborate on their labels and get input from Automate on what labels are likely to be inaccurate and should be reviewed.

  • What languages does Automate work in?

    English is available for all models in Automate, including custom models. More languages will be available later, let us know if you have a high priority language you need supported.

  • What format data does the data need to be to upload it?

    Creating custom models requires semi-structured data, so for now we support spreadsheet/ tabular data such as CSV. Our services team can help get your data ready for Automate or custom modeling if it’s not already tabular.

  • Who do I need to hire to make a custom model in Automate?

    Your existing team can work together to make their own machine learning model. Whether you’re a team of 1 or dozens, Automate allows you to share datasets, models, and training cycles so you can collaborate on the right solution together.

  • How long does it take to train a model?

    Most of the time training a model is spent labeling data – a team of 3-5 people can create all the labels you’ll need in a few hours. Once a training cycle is kicked off it takes about 20 minutes for the model to train and deploy. Most people can build and deploy a model in less than 48 hours.

  • How many pieces of training data is needed to be labeled?

    It varies – Automate will use active learning to reduce the number of labels you need to get a high performance model. What performance you need to deploy your model to production may vary based on your use cases so with each training cycle we show you the model’s metrics. Many of our users can train their models to human level performance with only a few hundred training examples.

  • How much does it cost?

    A traditional in-house machine learning team can cost millions. Get started with Automate today for a fraction of the cost – click here to chat with us.

  • Do I need to be able to write code to be able to build these models?

    Not at all! Automate handles the hard machine learning ops and research work all you have to do is drag and drop your data, create some labels, hit the train button and let the magic happen. Getting results from your model is no more complicated. If you have developers on your team they can access the model via API for any high-scale integration you need, but this is optional.