# FlowMesh: A Service Fabric for Composable LLM Workflows

arXiv: https://arxiv.org/abs/2510.26913 · HTML: https://mlsanity.com/paper/2510.26913/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Oct 2025.

## Built on by (most cited first, 6 of 6)

- [Efficient LLM Serving for Agentic Workflows: A Data Systems Perspective](https://mlsanity.com/paper/2603.16104/) (2026)
- [GraphFlow: A Graph-Based Workflow Management for Efficient LLM-Agent Serving](https://mlsanity.com/paper/2605.22566/) (2026)
- [FlowHijack: A Dynamics-Aware Backdoor Attack on Flow-Matching Vision-Language-Action Models](https://mlsanity.com/paper/2604.09651/) (2026)
- [TOPAS: Workflow-Aware Prefix-State Scheduling for Multi-Agent LLM Serving](https://mlsanity.com/paper/2608.25523/) (2026)
- [HeraSys: Collaborative Serving of Multiple LLM Workflows via Fine-Grained End-to-End Optimization](https://mlsanity.com/paper/2607.22578/) (2026)
- [DynBranch: Speculative Subgraph Reuse for Dynamic Agentic LLM Serving](https://mlsanity.com/paper/2609.31047/) (2026)

## Abstract (arXiv)

AI deployment increasingly resembles a pipeline of data transformation, fine-tuning, and agent interactions rather than a monolithic LLM job; recent examples include RLHF/RLAIF training and agentic workflows. To cope with this shift, we propose FlowMesh, a multi-tenant service fabric that executes and optimizes these workloads as one shared service instead of isolated pipelines. It decomposes workflows into fine-grained operators with recorded lineage, enabling de-duplication of work across users and batching requests on the same hardware while preserving per-workflow provenance. A global control plane maintains a cluster-wide pool of ready operators and uses a single utility function to pick both the batch and the worker, balancing throughput, cost, and data locality on heterogeneous GPUs. The data plane is an elastic fleet of stateless workers backed by a content-addressable store, enabling rapid, automatic scale-out, safe retry after preemption, and portability across managed clusters such as Kubernetes and geo-distributed GPU marketplaces such as Vast.ai. Compared with baseline solutions, FlowMesh achieves up to 3.8x cost reduction and 2.0x lower energy usage, provides a similar or better latency profile, and remains efficient under dynamic and failure-prone conditions.