做视频生成或世界模型的研究者终于有了正经的评估工具——WBench 把视觉质量和控制能力分开测，看完你会明白为什么很多漂亮视频其实不能当世界模型用。
LongCat 发布了 WBench，一个用于测试视频世界模型的基准，将测试重点从视觉质量转向控制、多轮记忆、指令遵循和物理合理性。WBench 包含 289 个案例、1058 次交互、20 个模型、5 个维度和 22 个自动指标，覆盖导航、主体动作、事件编辑、视角切换等。测试发现，没有模型在所有维度上占优，视觉质量与控制能力几乎无关。WBench 的设计将世界设置与用户动作分离，帮助研究者定位失败原因。这标志着视频世界模型评估从“视频好不好看”转向“模型能否维持可控世界”。
Most video models look better than they understand…
Most video models look better than they understand and Video quality is only the easiest thing to notice.
LongCat just released WBench, it turned video world model testing from a beauty contest into a stress test for control, multi-turn memory, instruction-following, and physical plausibility.
It exposed the gap between beautiful video generation and controllable world simulation.
A pretty clip is not enough, because a usable world model must keep the same scene, obey later actions, move the camera correctly, preserve objects, and avoid impossible cause-and-effect.
WBench tests this with 289 cases, 1,058 interaction turns, 20 models, 5 dimensions, and 22 automatic metrics, covering navigation, subject actions, event edits, perspective switches, and both viewpoints.
Across all those 20 evaluated models, the paper finds that no model dominates all dimensions, which means current systems have not yet merged high-quality rendering, reliable control, long-horizon memory, and physical rule-following into one stable capability.
Its design separates the world setup from the user action, so researchers can identify whether a failure comes from weak rendering, poor scene setup, bad control, lost state, or broken physics.
Navigation has near-zero connection with visual quality, consistency, or physics, meaning a model can look strong while still failing to move on command.
The key shift: stop asking only “does the video look good?” and start asking “can the model keep a controllable world alive across many turns?”
🧵 1.