PlexRL: Cluster-Level Orchestration of Serviceized LLM Execution for RLVR

• URL: http://arxiv.org/abs/2605.20863v1
• Date: Wed, 20 May 2026 07:55:06 GMT
• Title: PlexRL: Cluster-Level Orchestration of Serviceized LLM Execution for RLVR
• Abstract summary: We present PlexRL, a cluster-level runtime for multiplexing unified LLM services across RLVR jobs.<n>Our implementation and evaluations demonstrate that PlexRL significantly improves effective cluster capacity and reduces user GPU hour cost by maximum 37.58%.
• Score: 33.08614139144317
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning with verifiable rewards (RLVR) has recently unlocked strong reasoning capabilities in large language models (LLMs), triggering rapid exploration of new algorithms and data. However, RLVR training is notoriously inefficient: long-tailed rollouts, tool-induced stalls, and asymmetric resource requirements between rollout and training introduce substantial idle time that cannot be eliminated by job-local optimizations such as synchronous pipelining, asynchronous rollout, or colocated execution. We argue that this inefficiency is structural. While idle gaps are unavoidable within individual RLVR jobs, they are largely anti-correlated across jobs and therefore exploitable at the cluster level. Leveraging this observation, we present PlexRL, a cluster-level runtime for multiplexing unified LLM services across RLVR jobs. By centrally managing model placement, state transitions, and function-level scheduling under strict affinity constraints, PlexRL time-slices LLM execution across jobs to fill otherwise idle periods without expensive model migration. Our implementation and evaluations demonstrate that PlexRL significantly improves effective cluster capacity and reduces user GPU hour cost by maximum 37.58% while preserving algorithmic flexibility and introducing minimal per-job overhead.

Related papers

• SortedRL: Accelerating RL Training for LLMs through Online Length-Aware Scheduling [54.276306194000405]
  We propose SortedRL, an online length-aware scheduling strategy for scaling reinforcement learning.<n>SortedRL reorders rollout samples based on output lengths, prioritizing short samples forming groups for early updates.<n>Experiments using LLaMA-3.1-8B and Qwen-2.5-32B on diverse tasks, including logical puzzles, show that SortedRL reduces RL training bubble ratios by over 50%.
  arXiv  Detail & Related papers  (2026-03-24T16:48:31Z)
• RollArt: Scaling Agentic RL Training via Disaggregated Infrastructure [49.88201789074532]
  Agentic Reinforcement Learning (RL) enables Large Language Models (LLMs) to perform autonomous decision-making and long-term planning.<n>We present RollArc, a distributed system designed to maximize throughput for multi-task agentic RL on disaggregated infrastructure.
  arXiv  Detail & Related papers  (2025-12-27T11:14:23Z)
• Role-Based Fault Tolerance System for LLM RL Post-Training [24.163081819380647]
  RL post-training interleaves training and inference workloads, exposing the system to faults from both sides.<n>Our key insight is role-based fault isolation so the failure in one machine does not affect the others.<n>We present RobustRL, the first comprehensive robust system to handle GPU machine errors for RL post-training Effective Training Time Ratio improvement.
  arXiv  Detail & Related papers  (2025-12-27T06:30:18Z)
• RLBoost: Harvesting Preemptible Resources for Cost-Efficient Reinforcement Learning on LLMs [48.94639777633359]
  We present RLBoost, a systematic solution for cost-efficient RL training that harvests preemptible GPU resources.<n> RLBoost increases training throughput by 1.51x-1.97x while improving cost efficiency by 28%-49% compared to using only on-demand GPU resources.
  arXiv  Detail & Related papers  (2025-10-22T04:19:37Z)
• MACE: A Hybrid LLM Serving System with Colocated SLO-aware Continuous Retraining Alignment [14.392166280035122]
  Large language models (LLMs) deployed on edge servers are increasingly used in latency-sensitive applications such as personalized assistants, recommendation, and content moderation.<n>Existing retraining strategies either delay model updates, over-commit resources to retraining, or overlook iteration-level retraining granularity.<n>We propose MACE, a hybrid LLM system that colocates concurrent inference (prefill, decode) and fine-tuning, with intelligent memory management to maximize task performance while promising inference throughput.
  arXiv  Detail & Related papers  (2025-09-28T18:45:28Z)
• AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning [23.24949857136035]
  Reinforcement learning (RL) has become a dominant paradigm for training large language models (LLMs)<n>We present AReaL, a fully asynchronous RL system that completely decouples generation from training.
  arXiv  Detail & Related papers  (2025-05-30T07:18:25Z)
• StreamRL: Scalable, Heterogeneous, and Elastic RL for LLMs with Disaggregated Stream Generation [55.75008325187133]
  Reinforcement learning (RL) has become the core post-training technique for large language models (LLMs)<n>StreamRL is designed with disaggregation from first principles to address two types of performance bottlenecks.<n> Experiments show that StreamRL improves throughput by up to 2.66x compared to existing state-of-the-art systems.
  arXiv  Detail & Related papers  (2025-04-22T14:19:06Z)
• ArCHer: Training Language Model Agents via Hierarchical Multi-Turn RL [80.10358123795946]
  We develop a framework for building multi-turn RL algorithms for fine-tuning large language models. Our framework adopts a hierarchical RL approach and runs two RL algorithms in parallel. Empirically, we find that ArCHer significantly improves efficiency and performance on agent tasks.
  arXiv  Detail & Related papers  (2024-02-29T18:45:56Z)
• Efficient Parallel Reinforcement Learning Framework using the Reactor Model [2.190190313041532]
  Reinforcement Learning (RL) frameworks are essential for mapping RL workloads to multiple computational resources. Existing frameworks, such as Ray, are not managing this orchestration efficiently. We have proposed a solution implementing the reactor model, which enforces a set of actors to have a fixed communication pattern.
  arXiv  Detail & Related papers  (2023-12-07T21:19:57Z)
• Pollux: Co-adaptive Cluster Scheduling for Goodput-Optimized Deep Learning [61.29990368322931]
  Pollux improves scheduling performance in deep learning (DL) clusters by adaptively co-optimizing inter-dependent factors. Pollux reduces average job completion times by 37-50% relative to state-of-the-art DL schedulers.
  arXiv  Detail & Related papers  (2020-08-27T16:56:48Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.