{"version":"2.1.305","anchor":"workflow-parallel-and-pipeline-track-per-slot-state","canonical_anchor":"workflow-parallel-and-pipeline-track-per-slot-state","heading":"Workflow parallel() and pipeline() keep separate state for each item","tier":"internal","area":"Workflows","scope":"individual","heads_up":false,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.305\/e\/workflow-parallel-and-pipeline-track-per-slot-state","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.305","markdown":"### Workflow `parallel()` and `pipeline()` keep separate state for each item\n\nEach item run by the workflow `parallel()` and `pipeline()` helpers now keeps its own state, which is merged back afterwards\n\n**Unclear.** It is not clear what this changes in how workflows behave.\n\n**What**\n\nThe `parallel()` and `pipeline()` helpers in workflows now run each item in its own slot. A slot is a separate space that keeps that item's state apart from the others. When the items finish, their state is merged back together. Before, items ran directly with no per-item state.\n\n**Why**\n\nThis may affect the order in which workflow agents run or how their results are traced back to them.\n\n- Area: Workflows\n- Tier: Under the hood\n- Useful: 3\/5\n- Signal: 3\/5\n- Scope: individual\n- Heads-up: no"}