Hedra 推出 Model Factory，为机器人等行业提供定制模型训练服务
Hedra 把自动跑实验的智能体和 GPU 打包成服务，制造和机器人公司拿传感器数据来就能换回专属模型，部署后还会继续收集数据迭代。比起自己签 GPU 长约或养研究团队，这条路便宜不少，做具身智能和工厂自动化的团队可以关注。
Hedra labs 推出 Model Factory 服务，面向机器人、制造、生物和防御领域的团队提供定制模型训练。客户带来自己设备的传感器数据和仿真环境，Hedra 提供 GPU、运行训练实验的 AI 智能体，以及负责引导方向的研究人员。
客户最终拿到的是针对自家硬件训练的模型，无需签订长期 GPU 租约，也不用组建 AI 研究团队。模型部署后循环仍在继续：已上线的模型持续收集数据，在线评估系统对模型打分，结果再输入下一轮实验迭代，形成持续自我改进的闭环。
文中观点认为，持续自我改进（RSI）式的训练循环不再是前沿实验室的专属能力。这种服务模式让没有 AI 研究资源的制造和机器人企业也能获得持续进化的专用模型，与自建 GPU 集群或外包一次性模型训练的方式形成区别。
The frontier labs no longer own RSI loops.
With its new Model Factory, @hedra_labs is serving practically RSI to robotics, manufacturing, biology, and defense teams.
> A robotics or factory company brings its machines’ sensor data and a simulator of its setup, and Hedra brings GPUs, AI agents that run training experiments, and researchers who steer them.
> The customer leaves with a model built for its own hardware, without buying GPUs on long contracts or hiring an AI research team.
> The loop keeps going after deployment. The shipped model collects data, online evals score it, and that feeds the next round of experiments.
This report is categorized under AITOP's "Products" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include Hedra, 智能体, 机器人, Model Factory.
Source attribution: curated from rohanpaul_ai (@rohanpaul_ai).
