# X-World 1.6: Building Controllable Virtual Cities with World Models

DevFeed: [X-World 1.6: Building Controllable Virtual Cities with World Models](<https://devfeed.tech/articles/x-world-1-6-building-controllable-virtual-cities-with-world-models-115058.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.00596>)

Author: XPeng, :, Boyang Wang, Boyi Sun, Chaoda Zheng, Dongbin Zhang, Guangfeng Jiang, Hanlong Chen, Hongbin Lin, Jianlei Zheng, Jinhao Deng, Kai Yang, Kangjie Chen, Linkun Xu, Minqi Wang, Puyi Wang, Ronglian

Published: 2026-11-03T05:00:00Z

Content type: release

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>)

Topics: [World models](<https://devfeed.tech/topics/world-models.md>), [autonomous vehicles](<https://devfeed.tech/topics/autonomous-vehicles.md>), [post-training](<https://devfeed.tech/topics/post-training.md>)

Tags: [autonomous-driving](<https://devfeed.tech/tags/autonomous-driving.md>), [consistency](<https://devfeed.tech/tags/consistency.md>), [control](<https://devfeed.tech/tags/control.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [efficiency](<https://devfeed.tech/tags/efficiency.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [future](<https://devfeed.tech/tags/future.md>), [image](<https://devfeed.tech/tags/image.md>), [lights](<https://devfeed.tech/tags/lights.md>), [post-training](<https://devfeed.tech/tags/post-training.md>), [quality](<https://devfeed.tech/tags/quality.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [temporal](<https://devfeed.tech/tags/temporal.md>), [traffic](<https://devfeed.tech/tags/traffic.md>), [world-models](<https://devfeed.tech/tags/world-models.md>)

## AI overview

X-World 1.6 aims to make generative driving simulation more controllable and reliable during extended interactions. Its base model adds finer control over traffic lights and road markings and improves spatial consistency across cameras. Virtual City models use varied 2D and 3D references to build scenarios, including multi-camera future video generation from a single image. An enhanced post-training framework supports causal few-step generation, with reported gains in temporal coherence, quality over longer rollouts, and alignment with human preferences. A dedicated evaluation suite is also introduced. The abstract describes these capabilities but gives no quantitative results or implementation details.

## Source excerpt

arXiv:2611.00596v1 Announce Type: new Abstract: Deploying generative world models for end-to-end closed-loop evaluation and policy learning requires precise scenario control, consistent observations, and sustained generation quality during extended interactions. This report presents \emph{X-World~1.6}, advancing X-World~1.0 toward a more reliable and versatile driving simulator. The upgraded base model offers finer-grained control over traffic semantics, including traffic lights and road markings, together with improved spatial consistency across cameras. Virtual City capabilities broaden scenario construction through a family of models conditioned on diverse 2D and 3D references, including support for generating multi-camera future videos from a single image. An enhanced post-training framework enables efficient causal, few-step generation with stronger temporal coherence, better quality retention over longer rollouts, and improved alignment with human preferences. We also introduce a dedicated evaluation suite to support systematic assessment and guide model iteration. Together, these upgrades expand the scenarios X-World can simulate and improve the efficiency and reliability of its generated observations, supporting broader use in autonomous-driving development.