Tabnine defaults to a local configuration. Models are downloaded to your local development machine only - the code never leaves your machine.
Tabnine offers a cloud-based configuration, which allows for usage of more robust models that cannot be run locally on a development machine. These models are run on GPU servers, and require that you Opt-in before they can be used. We do not use your code to train our models. All information sent to our servers is ephemeral, only used to compute your prediction, and the queries themselves are not saved in our system after we’ve fulfilled them.
The Tabnine plugin gives you code completion suggestions based on a model trained on millions of open source programs, which are then modified based on the code you are currently working with. Tabnine uses the context of the code you are writing as a required input. With the default Tabnine configuration, our models are downloaded to your local machine in order to analyze your context in real-time while you are editing your code. Your code never leaves your machine.
We do not use your code to train our models. All information sent to our servers is ephemeral, is only used to compute your prediction, and the queries themselves are not saved in our system after they have been fulfilled.
Tabnine has an optional Cloud configuration. The cloud configuration serves our models from GPU-powered servers, letting us utilize the computing power we need to employ more powerful analysis models that cannot efficiently run on a development machine. If you choose to opt-in to Tabnine Cloud, Tabnine will send blocks of code from your edited files to our server, allowing us to provide deep completion suggestions. These blocks of code will never be stored - they are used to calculate predictions, and then immediately discarded.
Your code’s privacy is our highest priority. We at Tabnine want to reassure you that your code stays private when using our products. If you still have questions or reservations, please contact email@example.com