A world model is an AI system that learns a predictive representation of how an environment changes over time. It can use observations, actions, or language to estimate future states and possible outcomes before an agent acts. World models support planning in model-based reinforcement learning, robotics, embodied AI, autonomous driving, and interactive simulation. This site catalogs these systems with model records, research papers, benchmark context, and source links. Start with the world models database, compare systems on the leaderboard, or read the Performance Index methodology.
Static HTML snapshot of the model record for crawlers and no-JS readers.
| Attribute | Value |
|---|---|
| Model | RELIC |
| Lab / Organization | Adobe |
| Category | Generative World Model |
| Subtype | Interactive Video World Model with Long-Horizon Memory |
| World Model Type | Memory-aware interactive video world model |
| Primary Domain | Interactive Video |
| Architecture | Autoregressive video diffusion with memory-aware latent cache |
| Modality | Image + Text + Camera Actions → Interactive Video |
| Training Method | Autoregressive video-diffusion distillation with memory-aware rollout |
| Status | emerging |
| Year | 2025 |
Main editorial body preserved directly in static HTML.
RELIC is an Adobe Research project presented as an interactive video world model with long-horizon memory. Given an image and text description, it supports memory-aware real-time exploration while retaining spatial context through compressed historical latent tokens.
RELIC is a memory-aware interactive video world model developed by Adobe Research in 2025 for interactive video.
Short extractable facts for answer engines and no-JS readers.
| Signal | Value |
|---|---|
| Definition | RELIC is a memory-aware interactive video world model developed by Adobe Research in 2025 for interactive video. |
| Short Description | Adobe Research's interactive video world model combining real-time exploration, long-horizon spatial memory, and user control. |
| Benchmark Rows | 0 |
| FAQ Entries | 1 |
| Related Models | 2 |
| Related Guides | 0 |
| Related Research Topics | 0 |
| Last Updated | 2026-07-19 |
Key capabilities associated with this model.
Representative applications attached to this model record.
Balanced assessment surfaced in static HTML.
Primary references preserved in static HTML for citation extraction.
| Reference | Link |
|---|---|
| Adobe Research, 2025. RELIC: Interactive Video World Model with Long-Horizon Memory. | Open source |
FAQ answers rendered directly into static HTML for extractable responses.
Its project page explicitly presents RELIC as an interactive video world model with long-horizon memory.
Short extractable summary preserved directly in static HTML.
Editorial provenance and refresh policy preserved directly in static HTML.
Published by world-models.io editorial board.
Lead editor Tyler D. - Technical editor, methodology and benchmark analysis.
This model page synthesizes primary papers, official model pages, benchmark evidence, and related world-models.io context into a reference resource.
Each editorial page is assembled from primary sources, normalized into extractable summaries, checked for factual drift, and reviewed before publication or major refreshes. Last reviewed: 2026-07-19.
Pages are refreshed when a new paper, benchmark, release, architecture update, or stronger primary source materially changes the answer a reader or AI system should retrieve.
Each page links back to relevant primary sources and keeps a stable canonical URL so readers can verify claims, trace context, and reference the most up-to-date version. See the editorial policy.
Primary model and lab sources embedded in static HTML.