Chimera: Latency- and Performance-Aware Multi-agent Serving for Heterogeneous LLMs

• URL: http://arxiv.org/abs/2603.22206v1
• Date: Mon, 23 Mar 2026 17:01:42 GMT
• Title: Chimera: Latency- and Performance-Aware Multi-agent Serving for Heterogeneous LLMs
• Abstract summary: We present Chimera, a predictive scheduling system for multi-agent workflow serving on heterogeneous LLM clusters.<n>Chime traces the best latency, reducing end-to-end latency by 1.2--2.4$times$ and improving task performance by 8.0-9.5 percentage points on average over competitive baselines.
• Score: 62.17306142810532
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Multi-agent applications often execute complex tasks as multi-stage workflows, where each stage is an LLM call whose output becomes part of context for subsequent steps. Existing LLM serving systems largely assume homogeneous clusters with identical model replicas. This design overlooks the potential of heterogeneous deployments, where models of different sizes and capabilities enable finer trade-offs between latency and performance. However, heterogeneity introduces new challenges in scheduling across models with diverse throughput and performance. We present Chimera, a predictive scheduling system for multi-agent workflow serving on heterogeneous LLM clusters that jointly improves end-to-end latency and task performance. Chimera applies semantic routing to estimate per-model confidence scores for each request, predicts the total remaining output length of the workflow, and estimates per-model congestion using in-flight predicted token volumes for load balancing. We evaluate Chimera on representative agentic workflows for code generation and math reasoning using multiple heterogeneous LLM configurations. Across comparable settings, Chimera traces the best latency-performance frontier, reducing end-to-end latency by 1.2--2.4$\times$ and improving task performance by 8.0-9.5 percentage points on average over competitive baselines including vLLM.

Related papers

• End-to-End Latency-Minimizing and Load-Balanced Request Scheduling for Edge LLM Inference in Agentic AI Services [13.04369901641143]
  Large language model (LLM)-powered agentic AI services increasingly demand low-latency inference.<n>We develop a cross-slot inference model that captures transmission, prefill, redistribution-level decoding, and key-value cache evolution for each diverse request.<n>We propose the LYREO approach that transforms the long-term load-balancing constraint via Lyapunov optimization and employs reward with sequencebased return prediction.
  arXiv  Detail & Related papers  (2026-09-15T13:50:54Z)
• ASAP: Agent-System Co-Design for Wall-Clock-Centered Auto HPO Research for ML Experiments [45.577182866886126]
  We present ASAP, an agent-system co-design for HPO.<n>We show how ASAP integrates a diverse pool of inductive-biased generalizations and to select among their proposals each round.<n>We also show how ASAP re-architects the loop to reduce end-to-end wall-clock while preserving regret quality.
  arXiv  Detail & Related papers  (2026-06-23T22:00:26Z)
• Toward Reliable Design of LLM-Enabled Agentic Workflows: Optimizing Latency-Reliability-Cost Tradeoffs [22.661656301757663]
  This paper analyzes the fundamental tradeoffs between latency, reliability, and cost in LLM-enabled agentic.<n>We introduce performance models for both LLM and non-LLM agents that capture the relationship between computational effort and output quality.<n>Main results include a water-filling token allocation policy and characterizations of optimal workflow reliability in terms of shadow prices.
  arXiv  Detail & Related papers  (2026-04-21T23:09:16Z)
• Parallelization Strategies for Dense LLM Deployment: Navigating Through Application-Specific Tradeoffs and Bottlenecks [6.447439020088275]
  Breakthroughs in AI have fueled an explosion of large language model (LLM)-powered applications.<n>Dense models continue to dominate because of their strong ability, scalability, ease of fine-tuning, and versatility across diverse tasks.<n>This paper investigates the workloads of two representative dense LLMs, focusing in particular on intra-node parallelization schemes.
  arXiv  Detail & Related papers  (2026-03-05T21:33:24Z)
• ComAgent: Multi-LLM based Agentic AI Empowered Intelligent Wireless Networks [62.031889234230725]
  6G networks rely on complex cross-layer optimization.<n> manually translating high-level intents into mathematical formulations remains a bottleneck.<n>We present ComAgent, a multi-LLM agentic AI framework.
  arXiv  Detail & Related papers  (2026-01-27T13:43:59Z)
• Rethinking the Value of Multi-Agent Workflow: A Strong Single Agent Baseline [38.16649115214312]
  We show that a single agent can reach the performance of homogeneous with an efficiency advantage from KV cache reuse.<n>We propose an algorithm that automatically tailors for single-agent execution, reducing inference costs.
  arXiv  Detail & Related papers  (2026-01-18T08:16:09Z)
• Multi-Agent Tool-Integrated Policy Optimization [67.12841355267678]
  Large language models (LLMs) increasingly rely on multi-turn tool-integrated planning for knowledge-intensive and complex reasoning tasks.<n>Existing implementations typically rely on a single agent, but they suffer from limited context length and noisy tool responses.<n>No existing methods support effective reinforcement learning post-training of tool-integrated multi-agent frameworks.
  arXiv  Detail & Related papers  (2025-10-06T10:44:04Z)
• Difficulty-Aware Agentic Orchestration for Query-Specific Multi-Agent Workflows [5.376980683850103]
  We propose Difficulty-Aware Agentic Orchestration (DAAO), which generates query-specific multi-agent guided by predicted difficulty.<n>DAAO surpasses prior multi-agent systems in both accuracy and inference efficiency.
  arXiv  Detail & Related papers  (2025-09-14T03:57:43Z)
• Optimizing Sequential Multi-Step Tasks with Parallel LLM Agents [15.26802977779826]
  M1-Parallel is a framework that concurrently runs multiple multi-agent teams in parallel to uncover distinct solution paths.<n>We show that M1-Parallel with early termination achieves up to $2.2times$ speedup while preserving accuracy.<n>We further investigate strategies aimed at encouraging diverse execution plans but observe no additional performance gains over repeated sampling.
  arXiv  Detail & Related papers  (2025-07-11T18:09:22Z)
• Collab: Controlled Decoding using Mixture of Agents for LLM Alignment [90.6117569025754]
  Reinforcement learning from human feedback has emerged as an effective technique to align Large Language models.<n>Controlled Decoding provides a mechanism for aligning a model at inference time without retraining.<n>We propose a mixture of agent-based decoding strategies leveraging the existing off-the-shelf aligned LLM policies.
  arXiv  Detail & Related papers  (2025-03-27T17:34:25Z)
• MAMM-Refine: A Recipe for Improving Faithfulness in Generation with Multi-Agent Collaboration [63.31211701741323]
  We extend multi-agent multi-model reasoning to generation, specifically to improving faithfulness through refinement.<n>We design intrinsic evaluations for each subtask, with our findings indicating that both multi-agent (multiple instances) and multi-model (diverse LLM types) approaches benefit error detection and critiquing.<n>We consolidate these insights into a final "recipe" called Multi-Agent Multi-Model Refinement (MAMM-Refine), where multi-agent and multi-model collaboration significantly boosts performance.
  arXiv  Detail & Related papers  (2025-03-19T14:46:53Z)
• MALT: Improving Reasoning with Multi-Agent LLM Training [67.76186488361685]
  MALT (Multi-Agent LLM Training) is a novel post-training strategy that divides the reasoning process into generation, verification, and refinement steps.<n>On MATH, GSM8K, and CSQA, MALT surpasses the same baseline LLM with a relative improvement of 15.66%, 7.42%, and 9.40% respectively.
  arXiv  Detail & Related papers  (2024-12-02T19:30:36Z)
• Benchmarking Agentic Workflow Generation [80.74757493266057]
  We introduce WorfBench, a unified workflow generation benchmark with multi-faceted scenarios and intricate graph workflow structures.<n>We also present WorfEval, a systemic evaluation protocol utilizing subsequence and subgraph matching algorithms.<n>We observe that the generated can enhance downstream tasks, enabling them to achieve superior performance with less time during inference.
  arXiv  Detail & Related papers  (2024-10-10T12:41:19Z)
• Efficient Heterogeneous Large Language Model Decoding with Model-Attention Disaggregation [15.35494431928751]
  Transformer-based large language models (LLMs) exhibit impressive performance in generative tasks but also introduce significant challenges in real-world serving.<n>We introduce model-attention disaggregation to enhance the efficiency of LLM decoding.<n>We develop and deploy Lamina, an LLM inference system that incorporates model-attention disaggregation in a distributed heterogeneous cluster.
  arXiv  Detail & Related papers  (2024-05-03T02:15:15Z)

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.