Last date modified: 2026-Sep-01
Managing aiR for Review jobs
Use the aiR for Review Jobs tab to view and monitor job details at the instance and workspace levels.
- Monitoring in-flight jobs—see which jobs are queued, running, completed, or errored across every analysis type.
- Troubleshooting failures—quickly identify errored jobs and open the side panel to see failure reasons, document counts, and model or prompt versions.
- Administering aiR jobs at the instance level—review the most recent jobs across every workspace, including matter, workspace, and user context.
Additionally, authorized administrators can configure job size limits per user group at the instance level to help control and manage the size of jobs that users execute.
Setting aiR for Review job size limits
At the instance level, administrators can limit how many documents are included in an aiR for Review job per user group. This helps manage costs and prevents users from accidentally running large jobs.
The following points outline how job size limits are managed for user groups in the system:
- Limits are managed at the user group level, so every member of a group shares the same maximum job size.
- Different limits can be assigned to different groups.
- Multiple groups can be assigned the same job size limit
- Each group can only have one job size limit assigned at a time.
- If a user belongs to multiple groups with different limits, the highest limit among those groups applies when running a job.
To configure job size limits for user groups at the instance level:
- At the instance level, navigate to the Review Jobs page.
- Click Manage Job Limits above the table.
- On the Manage User Group Job Limits model, do the following:

- User Group(s)—select one or more user groups from the list. To remove a group from the field, click its corresponding circled X icon.
- Job Limit—enter the maximum number of documents (aiR Units) that users in that group can submit in a single job.
- To add more user groups and limits, click +Add Job Limit and repeat the previous step.
- To delete a job limit entry and its associated groups, click its corresponding trashcan icon.
- Click Save when done.
If users attempt to run a job that exceeds their document limit, an error appears for them showing the maximum size allowed, and the job does not execute.
Monitoring Review Jobs
There are two versions of the Jobs tab: one at the instance level, and one at the workspace level.
- Instance-level tab—System administrators can see the most recent 100 jobs across all workspaces in the instance on the Review Jobs page. It includes several extra columns to identify the workspace, matter, and client name connected to each job.
- Workspace-level tab—users at the workspace level can see all jobs within a single workspace on the aiR for Review Jobs tab.
If the aiR for Review Jobs tab message indicates aiR for Review is not currently available, check with your administrator. Your organization might not have an active contract for aiR for Review.
For information on managing tab permissions, see Permissions.
To open a project from the jobs list, click the Project Name link. The aiR for Review project opens for viewing and iteration editing.
To view details for a project, click its row to open the details panel. It displays attributes specific to the analysis type, so the fields displayed may vary.
Jobs list columns
- Analysis Type—analysis type of the job: Relevance, Key Documents, Issues, Confidential Business Information, or Custom.
- Project Name—name of the aiR for Review project associated with the job. Click the project name link to view project details.
- Job Status—current state of the job. The possible statuses are: Not Started, Queued, In Progress, Completed, Cancelling, and Errored.
- Doc Count—number of documents submitted for analysis
- User Name—name of the user who submitted the job.
- Matter Name (instance-level view only )—name of the matter associated with the job's workspace.
- Matter Number (instance-level view only )—number of the matter associated with the job's workspace.
- Workspace ID (instance-level view only )—ID of the job's workspace.
- Workspace Name (instance-level view only )—name of the job's workspace.
- Submitted Time—time the job was submitted for analysis.
- Completed Time—time the job successfully completed. If the job failed or was canceled early, this field is blank.
- Terminated Time—time the job stopped running, regardless of whether it was canceled, failed, or completed successfully.
Project details panel fields
Click a row to open the details panel for the project. It displays attributes specific to the analysis type, so the fields displayed may vary. The list below contains all possible fields. Only the fields specific to the project’s analysis type will display.
- [Project Name]—name of the project displays in the panel heading.
- Result Visibility—use this field to link and unlink analysis types to a job. Options include Hidden, Partially Hidden, and Visible. See Linking/Unlinking analysis types for more information. This option is not available for Custom analyses projects.
- Insights Successful—appears only for Custom analyses projects. Number of insights that generated successful outputs.
- Insights Errors—appears only for Custom analyses projects. Number of insights failed to generated outputs.
- Job Failure Reason—reason the job failed.
- Prompt Criteria Name—the name of the prompt criteria used by the job. If several jobs ran using the same prompt criteria, this name will be the same for those jobs.
- aiR for Model Version—the unique ID that identifies the underlying aiR system used to process each job. This ID does not map directly to a single large language model version used for the job.
- Prompt Version—the prompt criteria version associated with the job. For more information, see How prompt criteria versioning works.
- Docs Successful—number of documents that were successfully analyzed.
- Docs Errored—number of documents that encountered an error during analysis.
- Docs Pending—number of documents that are waiting to be analyzed.
- Docs Skipped—the number of documents that aiR did not return results for. This can happen for reasons such as canceling a job, network errors, and partial or complete job failures.
- Estimated Wait Time—estimated wait time for job to begin.
- Estimated Run Time—estimated wait time for job to complete.
The analysis type project information repeats from the Jobs list page.
Linking/Unlinking analysis types
When aiR for Review analysis has been run multiple times on the same document, each set of results is saved as part of a separate job. By default, when you look at a document's results, the results from the most recent analysis job are displayed. However, if you want to see the results from a previous job instead, you can link an older job's results to the document. Each set of results can be linked or unlinked at any time without losing any data.
For example, if you realize your current prompt criteria gives you less helpful results than a previous prompt criteria did, you can make the previous job's results visible. This immediately gives reviewers access to the old predictions without needing to re-run the old prompt criteria.
If you are viewing results from within the aiR for Review dashboard, the project version you select from the dashboard controls which job's results you see. If you are viewing results from other parts of Relativity, such as Review Center or the Document list, the job selected from the aiR for Review Jobs tab takes precedent. For more information on version selection in the dashboard, see How prompt criteria versions affect the Viewer.
To link (show) and unlink (hide) analysis types, do the following:
- Navigate to the aiR for Review Jobs tab.
- Click See details for the job.
- From the Result Visibility list, select one of the following:
- Visible—links all documents from this job so their results are visible in the Viewer. See aiR for Review Analysis for more information on aiR for Review analysis results in the Viewer.
- Partially Hidden—only some of the documents from this job show in the Viewer. For example, if a few documents from this job were later included in a different job, they might have that more recent job's results showing instead.
- Hidden—unlinks all documents from this job so their results are hidden in the Viewer. See aiR for Review Analysis for more information on aiR for Review analysis results in the Viewer.
- If you run a new job on documents that were part of a previous job, you may temporarily see both sets of results linked to those documents. The old results will be unlinked after the new job is complete.
- To avoid seeing doubled results, hide the previous result set using the aiR for Review Jobs tab.