{
  "id": 5123,
  "url": "https://arxiv.org/abs/2605.00528v2",
  "title": "SAGA: Workflow-Atomic Scheduling for AI Agent Inference on GPU Clusters",
  "summary": "AI agents execute tens to hundreds of chained LLM calls per task, yet GPU schedulers treat each call as independent, discarding gigabytes of intermediate state between steps and inflating end-to-end latency by 3-8x. We argue that this request-level abstraction is fundamentally mismatched to compound AI workloads, and propose a shift to program-level scheduling: treating the entire agent workflow (not individual inference calls) as the first-class schedulable unit. We present SAGA, a distributed ",
  "authors": "Dongxin Guo, Jikun Wu, Siu Ming Yiu",
  "category": "research",
  "topics": "agents-autonomy",
  "orgs": null,
  "regions": null,
  "published_at": "2026-05-01T09:05:28.000Z",
  "fetched_at": "2026-07-14T16:31:31.212Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/5123",
  "original_url": "https://arxiv.org/abs/2605.00528v2",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}