JoyNexus：面向服务的VLA模型多租户后训练

摘要

视觉-语言-动作（VLA）模型的后训练因模拟器、机器人本体和任务目标的多样性而至关重要。现有的计算服务（无论是直接租赁加速器还是提交批处理工作负载）通常为单个租户分配独占的GPU和CPU资源集。这种范式虽最大化客户灵活性，却迫使用户进行基础设施适配，且固定卡时的计费模式使短时或突发性工作负载既让租户成本高昂，又令服务提供商效率低下。为应对这些挑战，我们提出JoyNexus——一个面向多租户VLA监督微调、强化学习与评估的统一服务平台。JoyNexus将训练模型服务、推理模型服务与环境服务解耦，各自通过API访问，并依托驻留的共享基础模型（附租户专属槽位）运行。租户可直接调用高级语义API进行训练、模型部署与评估，或利用底层API及其分配端点组合自定义算法。多租户并发提交工作负载时，其动作模块、优化器、部署记录及策略版本保持隔离，服务由全局训练队列与推理队列调度。为提升多租户训练效率，JoyNexus针对共享兼容模型前缀的异构VLA数据模式引入分组批处理，使分组样本可复用单一共享主干前向传播。最后，我们通过工作负载仿真及实际具身场景中的分组批处理流水线评估JoyNexus。结果表明，与隔离的单租户执行相比，JoyNexus通过共享资源上的跨租户调度，减少了总GPU时间并提升了服务利用率。

English

The post-training of Vision-Language-Action (VLA) models is essential due to the diversity of simulators, robot embodiments, and task objectives. Existing compute services, whether offered as direct accelerator rental or batch-workload submission, typically allocate an exclusive set of GPU and CPU resources to a single tenant. While this paradigm maximizes client flexibility, it burdens users with infrastructure adaptation, and the fixed card-hour accounting model renders short or bursty workloads both expensive for tenants and inefficient for the service provider. To address these challenges, we present JoyNexus, a unified service for multi-tenant VLA supervised fine-tuning, reinforcement learning, and evaluation. JoyNexus decouples the Training Model Service, Inference Model Service, and Environment Service, each accessed through APIs and backed by resident shared base models with tenant-specific slots. Tenants can directly invoke high-level semantic APIs for training, rollout, and evaluation, or compose custom algorithms using lower-level APIs and their assigned endpoints. Multiple tenants submit workloads concurrently; their action modules, optimizers, rollout records, and policy versions remain isolated, and the service is scheduled by the global Training Queue and Inference Queue. To further improve multi-tenant training efficiency, JoyNexus introduces group batching for heterogeneous VLA data schemas that share a compatible model-facing prefix, enabling a single shared backbone forward pass over grouped samples. Finally, we evaluate JoyNexus through workload simulation and a group-batching pipeline in a realistic embodied scenario. Results show that, compared with isolated single-tenant execution, JoyNexus reduces aggregate GPU time and improves service utilization via cross-tenant scheduling on shared resources.