GLaaS Dags -> TReqs

The TReqs team is building tools for the future of ML development. Capture. Store. Control.

Capture what ran with roar. Store the record in GLaaS. Control what runs next with TReqs.

Together these components give you a complete record of how your models are actually built — and what gets to run next. Use any piece on its own.

When training requests are run on compute targets orchestrated on TReqs, the lineage gets auto-registered in GLaaS.

GLaaS includes key actions that connect your lineage context back to TReqs workflows, ensuring you can always decide what gets to run (and why) before launching compute jobs from small experiments to extra-large.

View Training Request

On a DAG page, click the 'View Training Request' button to open the training request in TReqs.

In TReqs, if you have permission to view the training request, you can see the request details:

• peer review and discussion
• compute target activity, logs, and execution history
• training request timeline and updates

This gives you a collaboration and execution view of the same lineage context represented in GLaaS.

Run in TReqs

Use this action to clone a DAG context into TReqs as a new workflow-backed training request.

In practice, this means you can launch work in your own TReqs org and project, while preserving the original context for perfect reproducibility. TReqs training requests are like pull requests for compute, allowing you to easily queue up past work to run again with the exact same data, code, and environment, pending team approval.