Wayve 依托 CoreWeave 构建自动驾驶模型持续训练与评估闭环
Wayve 是端到端自动驾驶的代表，这次讲了它怎么用 CoreWeave 把训练和路测评估跑成一个闭环，新模型必须赢过线上版本才能上线。做自动驾驶或大模型基础设施的人可以看看这套工程流程。
Wayve 与 CoreWeave 合作，将自动驾驶模型训练、跟踪与评估整合为持续循环流程。其 AI Driver 系统从 70 多个国家采集的真实道路数据中学习，在没有高精地图和人工编写规则的情况下应对路况变化。
每个新模型必须证明驾驶表现优于已部署在路上的版本，才能通过迭代。这一机制把模型开发与实际路测表现直接挂钩，避免只在离线指标上做优化。
对自动驾驶行业而言，端到端路线的落地越来越依赖大规模算力与数据闭环能力，CoreWeave 提供的 GPU 基础设施成为这类对比评估流程的支撑方。
Every new @wayve_ai model has to prove it drives better than the one already on the road.
Wayve’s AI Driver learns from real-world data collected in more than 70 countries to navigate changing conditions, without maps or hand-written rules. On CoreWeave, Wayve runs model training, tracking, and evaluation in a continuous loop, connecting new model development directly with how those models perform.
Building autonomous driving is a human pursuit. We're built for yours. https://t.co/8QkfClZx2O
This report is categorized under AITOP's "Industry" 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 Wayve, CoreWeave, 自动驾驶, 端到端.
Source attribution: curated from CoreWeave (@CoreWeave).
