> For the complete documentation index, see [llms.txt](https://docs.uptiq.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.uptiq.ai/document-ai/document-ai-1/test-dashboard.md).

# Test Dashboard

Test Dashboard is a QA framework for benchmarking extraction against ground-truth data. Upload a test document, define the expected extraction result, configure field-level assertions, then run the test against a processor and see exactly which fields matched and which didn't. Results roll up into fleet-wide accuracy metrics across document types.

In the UI, Test Dashboard is reachable only via the Home dashboard's Quick Links — it isn't in the left sidebar.

<figure><img src="/files/gBGWAf99Jnz7IWxKPHXS" alt=""><figcaption></figcaption></figure>

| Capability             | Description                                                                                                    | Limit                                                                                          |
| ---------------------- | -------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------- |
| Test documents         | Upload and manage test documents per document type; activate or deactivate                                     | —                                                                                              |
| Ground truth           | Define the expected extraction result for a test document                                                      | —                                                                                              |
| Assertions             | Configure field-level comparison rules using built-in operators                                                | Operator templates available per document type                                                 |
| Run a test             | Execute extraction against a processor and compare to ground truth                                             | Model is selectable per document type: GPT-4.1, GPT-5.1, or Gemini 3 (Gemini 3 is the default) |
| Live progress          | Stream test execution status as it runs                                                                        | Server-sent events                                                                             |
| Rerun                  | Re-execute a test with a different configuration                                                               | —                                                                                              |
| Fleet accuracy         | Weighted average accuracy across all document types and test cases                                             | —                                                                                              |
| Document-type coverage | See test coverage and accuracy per document type, and compare accuracy across models in a configuration matrix | Passed ≥95%, Review ≥85%, Failed <85%                                                          |

{% hint style="info" %}
Test Dashboard supports regression testing and CI/CD integration — rerun the same test suite after a processor or prompt change to confirm accuracy hasn't regressed.
{% endhint %}

### Workflow

1. **Add a test document** — upload a sample document for a specific document type.
2. **Define ground truth** — enter the expected extraction result for that document.
3. **Configure assertions** — set field-level comparison rules, using operator templates as a starting point.
4. **Run** — execute the extraction test against a processor. Progress streams live.
5. **Review** — see field-by-field comparison, which assertions passed or failed, and an accuracy score.
6. **Rerun** — re-execute with an updated configuration to check whether an accuracy issue has been resolved.
7. **Monitor fleet accuracy** — track weighted average accuracy across all document types and test cases over time.

Roles - Owner, Admin, Member, Viewer: browsing the full 115-type list, model selection, and the Export report / Run all tests / Upload test case controls are all present and enabled.

### Related pages

* [Document Extraction](/document-ai/document-ai-1/document-extraction.md) is the processor this dashboard benchmarks.
* [Document Types](/document-ai/document-ai-1/document-types.md) defines the document types and schemas each test case targets.
* [Workspace](/document-ai/document-ai-1/workspace.md) exposes a quick link into this dashboard for day-to-day QA work.


---

# 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.uptiq.ai/document-ai/document-ai-1/test-dashboard.md?ask=<question>&goal=<endgoal>
```

`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.