> For the complete documentation index, see [llms.txt](https://mariadb.com/docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://mariadb.com/docs/gridgain/gridgain-9/gridgain9-development/machine-learning.md).

# Machine Learning

Deploy machine learning models to a GridGain 9 cluster and run inference predictions using the GridGain ML engine.

GridGain 9 can host machine learning models on the cluster and run inference close to your data. The ML engine uses [Deep Java Library (DJL)](https://djl.ai/) as its inference runtime and supports PyTorch, TensorFlow, and ONNX Runtime models.

The pages in this section walk you through a complete example and describe model deployment and prediction options in detail:

* [Get Started With GridGain ML](/docs/gridgain/gridgain-9/gridgain9-development/machine-learning/get-started-with-gridgain-ml.md) — a working example of deploying and using a sentiment analysis model.
* [GridGain ML Model Deployment and Running Predictions](/docs/gridgain/gridgain-9/gridgain9-development/machine-learning/model-deployment-and-predictions.md) — model formats, deployment methods, prediction types, and advanced usage.

{% columns %}
{% column %}
{% content-ref url="/pages/3ivkmy2oGdkBFeq7X4Xk" %}
[Get Started with GridGain ML](/docs/gridgain/gridgain-9/gridgain9-development/machine-learning/get-started-with-gridgain-ml.md)
{% endcontent-ref %}
{% endcolumn %}

{% column %}
A complete, working example of deploying and using a sentiment analysis model with the GridGain 9 machine learning engine.
{% endcolumn %}
{% endcolumns %}

{% columns %}
{% column %}
{% content-ref url="/pages/Hk1rsrIh1BN4Dhghcq86" %}
[Model Deployment and Predictions](/docs/gridgain/gridgain-9/gridgain9-development/machine-learning/model-deployment-and-predictions.md)
{% endcontent-ref %}
{% endcolumn %}

{% column %}
Deploy machine learning models to a GridGain 9 cluster and run simple, batch, SQL-based, and colocated predictions, including advanced custom translators and compute jobs.
{% endcolumn %}
{% endcolumns %}