> For the complete documentation index, see [llms.txt](https://docs.mindee.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.mindee.com/models/live-test.md). # Live Test {% @supademo/embed url="" demoId="cmfmnqqnqgjnb39ozfh42qw7z" %} ## Overview The **Live Test** page lets you interactively test your model on real documents and visually inspect the extraction results in one click. It’s the perfect place to debug, validate, and refine your model’s behavior before moving to production. ## Key Features * **Side-by-side comparison**: Visually compare the raw document on the left with structured extraction results on the right, ideal for validating model behavior at a glance. * **Interactive document viewer**: Hover, click, and inspect highlighted fields directly on the image. Tooltips show extracted values and metadata in context. * **View raw JSON output**: Toggle to display the full JSON response returned by the API, allowing developers to inspect field names, values, coordinates, confidence scores, and nested structures. * **Instant visual feedback**: See where each field was extracted directly on the document. * **RAG Toggle**: If your have added documents into your RAG database, you can preview how RAG would influence the output. ## Accessing the Live Test Page The Live Test is available for all Mindee model types: extraction, crop, split, classification, .... From your home page on the Mindee platform, simply click on any model. The first page you see will be the Live Test page. You can always go back to the Live Test page by clicking on the left-hand model menu. ## Uploading a File for Testing 1. Upload a valid file (PDF or image) directly into the interface.\ You can click the "Upload a Document" button or drag and drop into the center of the window. 2. Mindee runs the current version of your model on the document. 3. The extracted fields appear on the right-hand side, alongside the image preview. 4. Each field is: * **highlighted on the document** with color-coded overlays, showcasing the polygon features that can be enabled in the data schema option. * listed in the right panel with values, field names, and optionally confidence scores if the [Confidence Score and Accuracy Boost](/extraction-models/optional-features/automation-confidence-score.md) feature is enabled.
Overview animation of Mindee Live View

By clicking the polygon, you can also see its corresponding field on the right side.

## Documents History When testing your model, you'll typically use several documents, and want to use those same documents for testing any changes made to the model. For this reason you can access the Documents History, where all your tests are stored. When you access a document in your history after having modified the Data Schema, you'll have a "Outdated Extraction" notice, along with a button to "Rerun Document".
Live Test - Rerun Document
### Deleting Documents You can delete documents and their results at any time. On the table view, select the documents you wish to delete, then click on the ":trash-can:" icon at the top of the page. {% hint style="info" %} Deleting a document is an irreversible action! The file and its corresponding data will be permanently removed from our system. {% endhint %} --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://docs.mindee.com/models/live-test.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.