--- title: "Submissions Overview" slug: "submissions-overview" description: "Learn how Hyperscience submissions group files, track lifecycle stages, support cases, and accept uploads through the application or API." updated: 2026-10-17T11:20:11Z published: 2026-10-17T11:20:11Z canonical: "help.hyperscience.ai/submissions-overview" --- > ## Documentation Index > Fetch the complete documentation index at: https://help.hyperscience.ai/llms.txt > Use this file to discover all available pages before exploring further. # Submissions Overview A submission is a set of files uploaded together for processing. The system interprets each file as a page, matches it to existing layouts, and groups it into documents based on these matches. The submission keeps the files, processing status, and results together. ## Submissions and cases Submissions and cases group content for different purposes: - A **submission** groups the files assigned to one processing record. - A **case** groups related content according to a business process, even when that content arrives in different submissions. For example, a customer might send documents for one mortgage application on two different dates. Hyperscience creates a separate submission for each arrival, but a case can group the content under one Case ID. For more information, see [Cases](/v44/docs/case-collation). ## Submission lifecycle A flow defines how Hyperscience processes a submission. A typical document processing flow moves a submission through the following lifecycle: 1. After the files are uploaded, Hyperscience creates the submission, assigns its Submission ID, and starts the selected flow. 2. The system prepares each supported source file for processing and creates its pages. 3. The system classifies the pages and groups them into documents according to the flow. - If Hyperscience cannot classify a page automatically, a keyer classifies it in a Document Classification Task. - Pages that are not assigned to a document remain unmatched. For more information, see [Matching Pages to Layouts](/v44/docs/matching-pages-to-layouts). 4. The system identifies and transcribes fields and tables. - When confidence is below the applicable Threshold, Hyperscience sends the work to Supervision. - The system can also send a sample of results to Quality Assurance (QA). For more information, see [Supervision and QA Introduction](/v44/docs/supervision-and-qa-introduction). 5. The system completes processing and makes the submission output available to downstream systems. ### Submission processing **Stage** and **Flow Run Status** describe different parts of this lifecycle. Use both values to understand whether processing is active and where the submission is in the flow. #### Stages The **Stage** value shows where a submission is in its flow. The Submissions table and the **Status** card on the Submission Output page show this value. The stages depend on the configured flow. For more information, see [Document Processing Flow](/v44/docs/document-processing-flow). | Stage | Meaning | | --- | --- | | Processing | Hyperscience is running an automated processing step. | | Manual Classification | A page requires classification in a Document Classification task. | | Manual Identification | A field or table requires manual identification. | | Manual Transcription | Content requires manual transcription. | | Flexible Extraction | Content requires Flexible Extraction supervision. | | Custom Supervision | Content requires a task defined by a custom flow. | | Complete | Hyperscience completed submission processing. | #### Flow Run Status **Flow Run Status** shows the state of the top-level flow run that processes the submission. It is separate from **Stage** and does not summarize the status of related flow runs. | Flow Run Status | Meaning | | --- | --- | | Running | The flow run is in progress. | | Completed | The flow run completed. | | Failed | The flow run failed. | | Cancelled | A user canceled the flow run. | If a flow run fails or is canceled, the submission keeps the last stage it reached. Use **Flow Run Status** to determine whether processing is running, completed, failed, or canceled. Use **Stage** to determine where processing is active or where it stopped. For more information, see [Flow Runs Page](/v44/docs/flow-runs-page). ## Accepted file types Hyperscience supports the following file types for submission processing: | File type | Processing notes | | --- | --- | | BMP | — | | DOC, DOCX | Hyperscience converts the file to PDF before it creates pages. The conversion can change the output. Use another supported format when output consistency is important. | | EML | Hyperscience can process the email and its attachments. You can turn off attachment processing. For more information, see [Process EML file attachments](/deployment/docs/processing-eml-files-attachments). | | GIF | For an animated GIF, Hyperscience extracts only the first frame. | | HEIC, HEIF | — | | HTM, HTML | Hyperscience converts the file to PDF before it creates pages. The conversion can change the output. Use another supported format when output consistency is important. | | JPEG, JPG | — | | MSG | Hyperscience can process the message and its attachments. | | PDF | PDFs with editable form fields are not supported. | | PNG | — | | TIFF | — | | TXT | TXT files are supported through the API, not through the application uploader. | | XLS, XLSX | Hyperscience converts the file to PDF before it creates pages. The conversion can change the output. Use another supported format when output consistency is important. | | XPS | — | | ZIP | Password-protected ZIP files are not supported. | ### Restrictions and processing notes - **Document conversion** — Hyperscience converts DOC, DOCX, XLS, XLSX, HTM, and HTML files to PDF before it creates pages. The conversion can change the output, so use another supported format when output consistency is important. - **Encryption and passwords** — Encrypted files and password-protected ZIP archives are not supported. - **Email attachments** — Hyperscience processes the attachments of EML and MSG files. You can turn off attachment processing for EML files. For more information, see [Process EML file attachments](/deployment/docs/processing-eml-files-attachments). - **Unaccepted file types** — Hyperscience does not create pages from files of other types. A submission created through the API with no accepted files fails with the 400 error “No valid documents supplied”. - **Training data** — You can upload any accepted file type as training data in Training Data Management (TDM). Training data imported from a ZIP file must contain only PNG, JPEG, TIFF, PDF, HEIC, or HEIF files; the import rejects archives that contain other file types. ## Create a submission You can create a submission by uploading files in the Hyperscience application or by sending them through the API. Input connections, such as folder and email listeners, also create submissions automatically. For more information, see [Overview of Integration Capabilities](/v44/docs/overview-of-integration-capabilities). ### Turn on Submission Upload To upload files in the application, the **Submission Upload** setting must be turned on: 1. Go to **Administration** > **System Settings**. 2. In the General section, click **Edit** on the **Submission Upload** card. 3. Set **Submission upload** to **Enabled**. 4. Click **Save**. ![Submission Upload settings dialog shows the Submission upload option enabled with Cancel and Save controls.](https://cdn.us.document360.io/87894cef-4958-4f3f-be6f-b75a78c82548/Images/Documentation/submission upload.png?sv=2026-02-06&spr=https&st=2026-11-04T07%3A10%3A30Z&se=2026-11-04T08%3A10%3A30Z&sr=c&sp=r&sig=bgBiL5PiGrPwcNTYiW63n9Lzfv1%2FbcifX%2BWklN%2BZghI%3D) When the setting is on, the Submissions page includes the **Create Submission** button. ![The Submissions page with the Create Submission button in the upper-right corner.](https://cdn.us.document360.io/87894cef-4958-4f3f-be6f-b75a78c82548/Images/Documentation/create_submission.png?sv=2026-02-06&spr=https&st=2026-11-04T07%3A10%3A30Z&se=2026-11-04T08%3A10%3A30Z&sr=c&sp=r&sig=bgBiL5PiGrPwcNTYiW63n9Lzfv1%2FbcifX%2BWklN%2BZghI%3D) ### Upload files in the application 1. Go to **Submissions** > **Submissions**. 2. Click **Create Submission**. 3. Add your files. Drag them into the upload area, or click the upload area and select the files. 4. If you upload multiple files, select how to submit them: - **Single Submission** creates one submission that contains all uploaded files. - **One Submission per file** creates a separate submission for each uploaded file. For example, uploading five files creates five submissions. 5. Click **Next**. 6. Select the flow that processes the submission from the **Flow** drop-down list. 7. Optionally, configure the submission: - Select a layout. This option applies to Semi-structured documents only. - Select **Add to an existing case** or **Create a new case** to add the submission to a case. For more information, see [Cases](/v44/docs/case-collation). - Turn on **Machine Extract Only** to process the submission without human review. This option does not change how the files are grouped into submissions. 8. Click **Upload**. > [!NOTE] > Automation and Accuracy > > Processing a submission with **Machine Extract Only** can reduce accuracy. For more information, see [Accuracy](/v44/docs/accuracy). Hyperscience creates the submission and lists it on the Submissions page. ![Upload Submissions dialog showing flows, case options, machine extraction toggle, and upload controls.](https://cdn.us.document360.io/87894cef-4958-4f3f-be6f-b75a78c82548/Images/Documentation/upload_subs.png?sv=2026-02-06&spr=https&st=2026-11-04T07%3A10%3A30Z&se=2026-11-04T08%3A10%3A30Z&sr=c&sp=r&sig=bgBiL5PiGrPwcNTYiW63n9Lzfv1%2FbcifX%2BWklN%2BZghI%3D) ### Create a submission through the API To create a submission through the API, authenticate your user and send a `POST` request to `/api/v5/submissions`. The files included in the same request belong to the same submission. The request has one required parameter, which depends on the encoding: - `file` for `multipart/form-data` and `application/x-www-form-urlencoded` encoding. The parameter accepts either a file as `multipart/form-data` or a string that defines the path of a file to retrieve and include in the submission. The prefix before `://` in the path defines the storage type where the file resides. For the supported storage types, see [Supported File Stores for Submission](https://docs.hyperscience.ai/#supported-file-stores-for-submission) in the API documentation. - `files` for `application/json` encoding. For each file, the parameter accepts one of the following JSON objects: - An object with the keys `file_content_base64`, which accepts Base64-encoded file data as a plain string or a data URI, and `filename`, which contains the file name. - An object with the key `file_url`, which accepts the URL of the file. The file must be one of the [accepted file types](/v44/docs/what-is-a-submission#accepted-file-types). You can add other parameters to the request, such as the flow, a layout, an external ID, or metadata. For more information, see the [Submission Creation API documentation](https://docs.hyperscience.ai/#submission-creation). ## Submission tracking Hyperscience assigns a Submission ID when it creates a submission. Use this ID to open the submission, monitor its processing, and retrieve its output. You can also provide the following tracking information when you create a submission through the API: - **External ID** — A unique identifier from an external system. Use it to search for the submission and retrieve its results. - **Metadata** — User-defined `JSON` data for external tracking. Metadata belongs to the submission, not to an individual document, and Hyperscience does not use it to process the submission. A submission also records its processing context, including the flow and Top-Level Flow run that process it. If your organization uses processing deadlines or SLAs, the submission can also have a processing deadline (`goal_time` in the API). For more information, see [Prioritizing Submissions](/v44/docs/prioritizing-submissions). ## Submission actions and data lifecycle Actions taken after creation affect a submission in different ways: | Action | Description | | --- | --- | | Cancel | Stops processing and cancels related flow runs and active Supervision or Quality Assurance tasks. The submission remains available and keeps the last stage it reached. You cannot undo a cancellation. | | Resubmit | Creates a new submission from the selected submission. It does not restart or replace the original submission. | | Delete | Deletes the submission. Deleting a submission from the Submissions table does not delete its Training Data Management records. | | PII data deletion | Removes potential PII according to the system-wide data-deletion policy. The submission can remain visible as **Data Deleted** or **Data Partially Deleted**. | For instructions to cancel, resubmit, or delete submissions, see [Submissions Page](/v44/docs/v44-navigating-the-submissions-page). ## Submissions page Use the Submissions page to find, monitor, and manage submissions. For information about its tabs, filters, actions, and columns, see [Submissions Page](/v44/docs/v44-navigating-the-submissions-page). ## Submission Output page The Submission Output page provides the content, results, and processing history associated with a submission. Depending on the flow and the processing outcome, it can include: - Source-file information and original files - Documents and document folders - Extracted fields and tables - Unmatched pages - Rejected documents - Cases associated with the submission's pages - External metadata - JSON output and transformed output - Processing activity For more information, see [Submission Output Page](/v44/docs/submission-output-page). The first step in Supervision. It is used to categorize and combine pages that were not classified by the machine. The confidence limit used to decide if a machine prediction should be sent for human review to ensure accuracy. A manual task that is created when the system’s confidence in a prediction is below the confidence threshold. Supervision allows a human to review and correct the output, ensuring data accuracy through human-in-the-loop input. Process that ensures the accuracy and reliability of system outputs. In Hyperscience, QA tasks allow users to review and correct errors in classification, identification, VLM extraction and transcription. Documents may be randomly sampled for QA from all processed data. The complete execution cycle of a specific flow within Hyperscience. Each flow run encompasses all steps from initiation to completion for a given submission, allowing you to monitor, troubleshoot, and manage document processing. A tool used to annotate, manage, import, and export training documents. It is also used to train models by working directly with the training data (“ground truth”) obtained from each document in the training set. **Automation** refers to the processing of data without the need for human intervention. **Accuracy** measures the effectiveness of the models based on the proportion of correct predictions out of all predictions made. It helps you understand how often the system correctly predicts values compared to the actual values that reached consensus during QA. Accuracy can be influenced by factors like imbalanced datasets or inconsistent annotations. The main flow that manages the end-to-end document processing, coordinating with subflows to handle specific components of the process. Learn more in [Flows](https://help.hyperscience.ai/latest/docs/flows). ## Related - [Submissions Page](/submissions-page.md) - [Submission Output Page](/submission-output-page.md) - [Document Output Page](/document-output-page.md)