DreamForge-World 0.1 Preview: A Low-Compute Real-Time Controllable World Model

arXiv 2026 43.9 system, method

TLDR

A low-compute, real-time controllable world model preview using adapted video backbone, achieving interactive simulation on consumer GPUs.

Reasoning

The paper presents a practical low-compute world model with real-time interactivity and broad capabilities, but it is a preview version with limited evaluation and not frontier-quality.

Read-first score

Read-first score 43.9, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 57.

Recency 6%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Topical relevance 29%

81.4

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Methodology quality 18%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=none

Reproducibility 18%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Citation impact 18%

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. cited_by_count=0

Citation velocity 12%

Citation velocity estimates citations per publication-year to reduce old-paper bias. velocity=0.00

Field roles

Frontier

Rank sensitivity

Stability: volatile; rank range: 472.

Keyword Scores

world model

world simulator

interactive world model

video world model

generative world model

world dynamics prediction

model-based reinforcement learning world model

Deep Analysis

Innovations

Low-compute adaptation of autoregressive video model (LongLive 1 from Wan2.1-T2V-1.3B) with residual action pathway for real-time interactive world simulation
Real-time controllable world model running at 14-15 FPS on a single consumer GPU (RTX 4090) with low memory footprint
Support for live keyboard and mouse control, multimodal initialization, mid-stream reprompting, dual-view operation, and minute-scale interactive rollouts
High cost-efficiency via leveraging open video backbones and targeted adaptation runs

Methodology

The system adapts the LongLive 1 autoregressive video stack (derived from Wan2.1-T2V-1.3B) by incorporating a residual action pathway inspired by the Matrix-Game family. It uses open video backbones and targeted adaptation runs to achieve low-compute, real-time interactive simulation on consumer GPUs.

Key Results

DreamForge-World 0.1 Preview achieves 14-15 FPS interactive rollouts at native 480p resolution on a single RTX 4090 with low memory footprint, supporting live keyboard/mouse control, multimodal initialization, mid-stream reprompting, dual-view operation, and minute-scale simulations.

Limitations

Not yet a memory-complete or frontier-quality world simulator
Preview system with limited capabilities compared to frontier-scale simulators

Resources

Paper