Back to News

Kunlun Wanwei releases Matrix-Game 4.0, real-time 480p/60FPS game model on single RTX 5090

#video generation#game model#real-time rendering#open source

On October 15, 2026, Kunlun Wanwei released Matrix-Game 4.0, claiming it is the world's first video-model game running on a single consumer GPU. It delivers 480p, 60 FPS real-time gameplay on an NVIDIA RTX 5090 with end-to-end latency below 150 ms, and the code and technical report have been open-sourced.

Coverage timeline

  1. 机器之心新闻资讯
    昆仑万维创始人周亚辉署名Matrix-Game 4.0 技术报告

    2026年10月15日， 昆仑万维正式发布世界上第一款运行在单张消费级显卡上的视频大模型游戏——Matrix-Game 4.0 。Matrix-Game 4.0 构建起了游戏世界模型的新范式： 由游戏引擎维护世界状态与交互规则，由视频生成模型实时呈现精细画面。 它把 480p、60 FPS 的实时游戏体验，装进了一张 NVIDIA RTX 5090；从你按下按键，到画面做出反应，延迟低于 150 毫秒。Matrix-Game 4.0 首次跑通了"代码生成游戏逻辑 → 结构画面提供结构约束 → 视频模型实时渲染"的完整闭环，让游戏原型开发、个性化定制与 UGC 从此拥有了真正可落地的技术路径。 项目已同步开源, 技术报告也已发布。据传，周亚辉亲自参与了Matrix Game4.0 游戏大模型研发工作”。 开源地址 及技术报告： Github： https://github.com/SkyworkAI/Matrix-Game.git Huggingface：https://huggingface.co/Skywork/Matrix-Game-4.0 项目网站： https://matrix-game-v4-0.github.io/page/ 技术报告：https://github.com/SkyworkAI/Matrix-Game/blob/main/Matrix-Game-4/docs/Matrix_Game_4_0.pdf 一、从一段代码到一个可玩的世界——Matrix-Game 4.0 的四块基石 1. 单卡实时：让"可玩的视频模型"走出实验室 在消费级显卡上实时生成游戏画面，曾是难以兼得的目标——视频生成模型对算力的胃口，长期把它锁在实验室的多卡集群里。 Matrix-Game 4.0 给出的答案是一组硬指标： 单张 NVIDIA RTX 5090，480p 分辨率下游戏以 60 FPS 运行 ；从玩家按键到首帧响应， 端到端延迟低于 150 毫秒 ；单步去噪的稳态生成速度达 83 FPS ， 显存峰值 18.5 GiB 。 这组数字意味着，视频模型驱动的游戏第一次具备了本地部署与实时交互的实际价值——它不再是一段离线渲染的演示视频，而是你能坐在电脑前真正上手操作的游戏。 2.Harness 新范式：让世界状态被真正"管理"起来 Matrix

Related stories

  • Qwen 3.8 Flash Next (125B) Runs on RTX 4090 at 100 Tokens/s

    A Hacker News post reports running Qwen 3.8 Flash Next, a 125B-parameter model, on consumer hardware (an RTX 4090) at 100 tokens per second. The post appears to be a hands-on engineering write-up, though details of the optimization techniques are not provided in the source.

    1 reports · 1 sources923 points · 421 comments

  • Mistral Large 4 preview: 1T-parameter open-weight model, weights due Oct 27

    Mistral AI released a public preview of Mistral Large 4 (ML4), a 1-trillion-parameter multimodal model with 49 billion active parameters, trained on 3,800 Nvidia Grace Blackwell GPUs in its European data centers. The open weights are promised by October 27. Early benchmarks show an Artificial Analysis Intelligence Index score of 38, comparable to DeepSeek V4.1 Flash, and the model is claimed to top any open model from the US or Europe.

    10 reports · 5 sources2033 points · 1210 comments

  • Alibaba CEO outlines machine intelligence vision, plans 5T-10T parameter Qwen models

    At the 2026 Yunqi Conference, Alibaba CEO Eddie Wu said machine thinking will scale to 1000 times human thinking, positioning AI models, chips, and cloud as the three pillars of the machine intelligence era. The company announced plans to train a new Qwen model with 5-10 trillion parameters, progress in recursive self-improvement (RSI), and the release of the Zhenwu V900 AI chip with 3x the compute of its predecessor.

    5 reports · 2 sources

  • Alibaba open-sources Qwen-Image-2.1, a 7B image model with unified generation and editing

    Alibaba's Qwen team open-sourced Qwen-Image-2.1, a 7B-parameter image generation model that unifies text-to-image and image editing in a single pipeline. It natively supports transparent image generation and accepts up to ten reference images, with enhanced local editing and portrait/product fidelity. According to the team, it scores 60.28 on the Qwen-Image-Bench, the highest among open-source models and ahead of Nano Banana 2.0 (59.82) and GPT Image 1.5 (59.65).

    3 reports · 3 sources735 points · 198 comments

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe