The Data Extraction API stores your processing runs (the input documents and the extracted outputs) as a product feature so you can view and manage your run history. You can delete runs, and runs expire automatically after a retention period that depends on your plan.
For plans where data retention is enabled, we also retain a copy of this data internally and use it to operate, improve, and develop our services, including to train and evaluate our models. This internal copy is separate from the run history you see and can delete. For plans where data retention isn’t enabled, your documents aren’t retained for these purposes.
Extracted content is also returned directly in the API response.
For full details, including the legal basis, how product storage differs from internal retention, and how to request deletion, see the Nutrient API Privacy Policy.
Also refer to the getting started guide, API overview, Data Extraction API overview, and pricing and security pages.