SwarmX deploys scheduling-specific neural predictors and a scheduler-agent framework that cut tail latency by up to 61.5% and double throughput versus prior schedulers in multi-agent code, research, and multimodal workloads.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.DC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
SwarmX: Agentic Scheduling for Low-Latency Agentic Systems
SwarmX deploys scheduling-specific neural predictors and a scheduler-agent framework that cut tail latency by up to 61.5% and double throughput versus prior schedulers in multi-agent code, research, and multimodal workloads.