SSyncDev

Matrix-Game 4.0

SyncDev model directory · Profile updated Nov 18, 2026

Compare with another model

Matrix-Game 4.0 is Skywork's open set of 5B real-time video renderers that turn a running game's frames into clean styled video, per scene.

Key specifications

Specification score
14.8
License
Apache-2.0

The specification score summarizes recorded features and limits. It is not a benchmark result.

Pricing conditions: No API pricing stated; open weights for self-hosting

Inputs and outputs

Recorded modality support shows what this model can accept and produce. An undocumented modality is shown as unknown rather than assumed to be unsupported.

Accepts

  • Video input
  • Image input

Produces

  • Video output

Capabilities

VisionOpen weights

What is Matrix-Game 4.0?

Earlier world models tried to be the whole game: physics, input handling and pictures in one network. Matrix-Game 4.0 gives up on that. A real, executable game decides what happens. The model only re-renders what the game produces into clean low-poly RGB video, frame-aligned with a separate HUD and sound layer.

Skywork released it on 15 October 2026 with three scene-specialised checkpoints, each with 5,001,016,512 BF16 parameters: Medieval Castle, Ancient Egypt and City Racing. It builds on Wan2.2-Fun-5B-Control. Inference code, a browser demo and the game source came out together.

What changes in practice

Because gameplay is deterministic code, the old failure modes of neural game engines (objects that vanish, rules that drift) move out of the model's hands. What remains is a rendering problem with a strict contract:

  • Output is 480×864 RGB. After start-up, each latent chunk maps to four frames.
  • The quality preset uses the Wan VAE with two denoising steps.
  • The realtime preset uses the TAEHV codec, a single step, an optimised single-GPU DiT and CUDA graphs.
  • The attention cache holds 21 latent frames including the identity sink. CFG is 1 and scheduler shift is 5.

Castle and Egypt use an independent identity sink; Racing uses the supplied image as frame zero. Skywork is explicit that these policies are not interchangeable.

Install notes

Item Detail
Hub repo Skywork/Matrix-Game-4.0
Pipeline Matrix-Game 4.0 inference source only; not a Diffusers drop-in
Shared parts T5 encoder, tokenizer and Wan2.2 VAE pulled from the base repo
Integrity Manifest downloader checks length and SHA256 of every file
Not included Training code and training data
Licence Apache-2.0

On 16 October Skywork added tuned inference for RTX 5080 and RTX 5070 Ti cards. The Hub repository predates the public release by three days.

Compared with version 3.0

Matrix-Game 3.0, from March 2026, was a streaming interactive world model with long-horizon memory. Version 4.0 narrows scope to rendering per scene in exchange for game-grade consistency. Choose 3.0 for open-ended world exploration and 4.0 when you need controllable, repeatable gameplay.

Is it useful to you?

Studios prototyping a neural art pass over grey-box levels should try it: feed your own capture and judge temporal stability over a long session. For new scenes you would need your own training, which this release does not cover. More from the lab is under Skywork's model listing, and related tools sit in the video-to-video directory.

How to evaluate Matrix-Game 4.0

Compare the model on the work you need to complete.

Tasks and inputs

Matrix-Game 4.0 is listed for Text-to-video, Video-to-video. It accepts video, image and produces video. Use the publisher’s input format and preprocessing for your chosen deployment.

Capacity and cost

Choose duration, resolution and frame rate together. Compare the cost of an accepted clip, including repeat generations.

Choosing a deployment

The weights are downloadable under Apache-2.0. Check runtime support and size the complete model and working context for your expected concurrency.

Frequently asked questions

It re-renders frames from an executable game into clean low-poly RGB video at 480×864; the game itself handles gameplay, physics and input.

These records share a directory category. Check the source card, output type and deployment requirements before treating a model as a replacement.

NVIDIA

Cosmos3-Nano-Sim-Auto

Cosmos3-Nano-Sim-Auto is NVIDIA's open causal driving simulator: real-time multi-camera video and LiDAR that react to an AV policy's planned path.

— context

NVIDIA

Cosmos3-Nano-Sim-Transfer

Cosmos3-Nano-Sim-Transfer is NVIDIA's open 16B model that renders simulator depth, segmentation or edge maps into photorealistic robot video in real time.

— context

NVIDIA

Cosmos3-Nano-Transfer-Auto

Cosmos3-Nano-Transfer-Auto is NVIDIA's open 16B model that generates synchronised multi-camera driving video and LiDAR from map controls and a prompt.

— context

Tencent

WeVisCut

WeVisCut is Tencent's 9B tool-calling agent that edits narration scripts into multi-minute videos through 12 editing tools and an editable Studio timeline.

— context

skywork

SkyReels-V3-V2V-14B

SkyReels-V3-V2V-14B is Skywork's 14B video-extension model that continues a clip for 5-30 seconds or cuts to a new shot, at 720p from a text instruction.

— context

AInVFX

AInVFX/SeedVR2_comfyUI

AInVFX/SeedVR2_comfyUI is a source-linked Hugging Face repository assigned to Video-to-video. Its recorded task interface accepts video and produces video.

— context