AIAny
Icon for item

Jeveration: System One Models Now Generate

Generates images in a single forward pass and handles image-to-image tasks and dense prediction with the same model. A shared random state coordinates image codes for global coherence; 512×512 generation takes about 78 ms on an H200.

Introduction

Fast parallel prediction usually suits tasks with one right answer; image generation is harder because many different images can fit the same prompt. The central insight here is to give every spatial position a shared random state: it coordinates parallel choices into a coherent image without iterative sampling.

Key Findings
  • A single checkpoint handles text-to-image, image-to-image generation, and dense prediction. This makes the work a test of whether a unified visual model can cover both creative and structured outputs, rather than optimizing only for image synthesis.
  • On an H200, 512×512 generation takes 77.8 ms in one pass and reaches 0.74 on GenEval. The paper reports it as faster than the compared unified models, with competitive image quality.
  • On NYUv2, the model reaches 67.7 mIoU for segmentation, close to dense-only models trained from the same backbone. Depth and surface-normal results show that sharing capacity across tasks can still have costs, particularly for fine local geometry.
How the Coordination Works

The model starts from a frozen vision-language backbone and uses sampled decision tokens for image generation or spatial queries for dense prediction. For generation, all image-code positions depend on the same random state. The authors first train a distribution over discrete image codes because direct image supervision tends to make the model ignore that state. This is the mechanism that lets a single parallel prediction retain variation while avoiding the incoherence of independent per-position choices.

Who Should Read It

A good fit for researchers exploring non-autoregressive image generation, unified vision models, or low-latency multimodal systems. Look elsewhere if you need evidence of production readiness or best-in-class dense prediction: the reported dense results are benchmark-specific, surface normals trail dedicated models, and the paper’s headline speed and quality figures do not by themselves establish performance across broader settings.

Information

More Items

Enables interactive world models to follow user-defined rules and preserve them across save-and-restore. An executable 3D rule graph tracks rules, dependencies, and object states, while constraining video generation to show their effects.

Scales verifiable gameplay data by having a Builder and visual and coding players co-develop web games. Progress-weighted training improves agent performance on both its own benchmark and an external game benchmark.

Generates videos directly in pixel space without a pretrained autoencoder or cascaded refinement stages. It can jointly produce RGB frames and disparity maps, helping generated scenes follow target camera motion.