Matrix-Game 4.0
The first video-model game playable in real time on a single consumer GPU: 480p/60 FPS, <150ms input-to-pixel latency on one RTX 5090, open-sourced with game code.
Why it matters
A coding agent writes the game logic, the engine maintains world state and interaction rules as an external constraint, and a video model renders rich visuals in real time — the first playable, open-source world-model game on consumer hardware.
What you could build with it
A small indie game studio could prototype full game concepts in days instead of months: describe the game in natural language, let a coding agent generate the logic and Matrix-Game render playable visuals in real time on a single RTX 5090, so players can feel the gameplay before any art pipeline exists.
Does it hold up?
Too early to judge — released today (Oct 15); no third-party builds or usage reports found yet. Claims (single RTX 5090, 83 FPS steady generation, 18.5 GiB VRAM peak) are vendor-reported; independent verification pending.
Built with Matrix-Game 4.0
- Kunlun Tech releases Matrix-Game 4.0, world's first consumer-GPU real-time world-model gamepress · Press coverage of the Oct 15 release: single RTX 5090, real-time playable world-model rendering, open-sourced game code.
Learn more
First spotted on github: source.
More AI models
OpenAI publishes 722 AI-generated math manuscripts from an unreleased modelOpenAI released 722 mathematical manuscripts in 372 result families to GitHub under Apache-2.0, produced by an…models · JEV 0.74StepFun Step-5-Preview — open BF16 weights releasedStepFun published the full BF16 weights for its flagship Step-5-Preview agentic model (600B-parameter sparse MoE, 27B…models · JEV 0.73ElevenLabs Eleven v4 + v4 Turbo: new emotive TTS architecture with a 100ms real-time variantElevenLabs released Eleven v4 and Eleven v4 Turbo on September 28: a new text-to-speech architecture with inline…models · JEV 0.72Mistral Large 4 (Le Chonk): 1.05T-param open-weight MoEMistral's new flagship: a 1.05T-parameter mixture-of-experts multimodal model with 49B active params and 1M-token…models · JEV 0.72VoiceStudio: open-source, fully-local ElevenLabs alternative with voice cloning, dubbing and dictation in 646 languagesVoiceStudio is an open-source, fully-local voice AI studio: voice cloning from a 3-second sample, voice design, video…models · JEV 0.71JetBrains Mellum 2.1: 12B MoE coding model trained for agentsJetBrains released Mellum 2.1, a 12B MoE model (2.5B active parameters, 128K context) under Apache 2.0, trained mainly…models · JEV 0.69
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.