---
title: "Genie 3 | Google DeepMind Real-Time World Model | world-models.io"
canonical: "https://world-models.io/en/models/genie-3/"
description: "Genie 3 is Google DeepMind's real-time interactive 3D world model that generates diverse environments from text prompts at 24fps."
author: "Tyler D."
generated: "2026-10-30"
---

New: the Timeline is live. Track world model releases, papers, and benchmark updates in real time.

The Knowledge Hub for AI World Models

# Genie 3

Genie 3 is Google DeepMind's general-purpose world model generating interactive 3D environments from text prompts in real time at 24fps with 720p resolution. Available publicly via Project Genie for Google AI Ultra subscribers.

robotics

model-based-rl

simulation

embodied-ai

## What is a world model?

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](https://world-models.io/en/world-models-database/), compare systems on the [leaderboard](https://world-models.io/en/leaderboard/), or read the [Performance Index methodology](https://world-models.io/en/methodology/performance-index/).

## Key Attributes

Static HTML snapshot of the model record for crawlers and no-JS readers.

| Attribute | Value |
| --- | --- |
| Model | Genie 3 |
| Lab / Organization | DeepMind |
| Category | Generative World Model |
| Subtype | Interactive 3D World Model |
| World Model Type | Real-time interactive 3D environment generation |
| Primary Domain | 3D World Generation |
| Architecture | Autoregressive latent world model with spatiotemporal transformer |
| Modality | Text → Interactive 3D Environment |
| Training Method | Large-scale internet video and 3D data pre-training with action-conditioned generation |
| Status | active |
| Year | 2025 |
| Performance Index | 89/100 (medium confidence, v1.1) |

## About Genie 3

Main editorial body preserved directly in static HTML.

Genie 3 is a general-purpose world model from Google DeepMind that generates an unprecedented diversity of interactive environments from text prompts. Users can navigate generated worlds in real time at 24 frames per second, with consistency maintained for several minutes at 720p resolution. Building on Genie 1 and Genie 2, this third generation dramatically expands the scope of generated worlds, from natural landscapes and ecosystems to architectural spaces and fantastical environments. The model demonstrates emergent understanding of physics, lighting, object permanence, and spatial relationships. Project Genie, the public product built on Genie 3, was made available to Google AI Ultra subscribers in January 2026.

Genie 3 is a real-time interactive 3d environment generation developed by Google DeepMind in 2025 for 3d world generation.

## Editorial Snapshot

Short extractable facts for answer engines and no-JS readers.

| Signal | Value |
| --- | --- |
| Definition | Genie 3 is a real-time interactive 3d environment generation developed by Google DeepMind in 2025 for 3d world generation. |
| Short Description | Google DeepMind's general-purpose world model that generates interactive 3D environments from text prompts in real time at 24fps. |
| Benchmark Rows | 2 |
| FAQ Entries | 2 |
| Related Models | 4 |
| Related Guides | 0 |
| Related Research Topics | 3 |
| Last Updated | 2026-04-10 |

## Notable Features

Key capabilities associated with this model.

- • Real-time interactive 3D world generation at 24fps
- • Text-to-world prompt interface
- • 720p resolution with multi-minute consistency
- • Emergent physics and lighting understanding
- • Publicly available via Project Genie

## Use Cases

Representative applications attached to this model record.

Game prototyping

Creative world-building

Architectural visualization

Training data generation

Embodied AI research

## Strengths and Limitations

Balanced assessment surfaced in static HTML.

### Strengths

- • First real-time interactive world model
- • Unprecedented environment diversity
- • Consumer-facing product
- • Strong physical plausibility

### Limitations

- • Consistency degrades after ~1 minute
- • Closed-source
- • Limited to Google AI Ultra subscribers
- • Cannot export generated worlds

## Benchmarks

Published benchmark evidence attached to this model record.

| Benchmark | Metric | Result | Source |
| --- | --- | --- | --- |
| Interactive World Rate | Frame Rate 24 fps | Real-time at 24fps | [Source](https://deepmind.google/blog/genie-3-a-new-frontier-for-world-models/) |
| World Consistency Horizon | Stable Rollout 60 seconds | Up to ~1 minute | [Source](https://www.gamesindustry.biz/googles-genie-3-world-models-start-to-break-down-after-around-a-minute) |

## References and Citations

Primary references preserved in static HTML for citation extraction.

| Reference | Link |
| --- | --- |
| Parker-Holder & Fruchter, 2025. Genie 3: A New Frontier for World Models. Google DeepMind Blog. | [Open source](https://arxiv.org/abs/1803.10122) |

## Related Models

Nearby models linked from the current editorial record.

| Model | Category | World Model Type | Index v1.1 |
| --- | --- | --- | --- |
| [Genie 2](https://world-models.io/en/models/genie-2/) | Generative World Model | Generative environment model | 79/100 |
| [Genie](https://world-models.io/en/models/genie-1/) | Generative World Model | Action-controllable generative model | 57/100 |
| [NVIDIA Cosmos](https://world-models.io/en/models/nvidia-cosmos/) | Foundation World Model | Video world foundation model | 87/100 |
| [OASIS](https://world-models.io/en/models/oasis/) | Generative World Model | Real-time playable world model | 66/100 |

## Direct Comparisons

Side-by-side comparisons already connected to this model.

| Comparison | Matchup | Summary |
| --- | --- | --- |
| [Genie 3 vs Genie 2](https://world-models.io/en/compare/genie-3-vs-genie-2/) | Genie 3 vs Genie 2 | Two generations of DeepMind's interactive world model. Genie 2 generates 3D environments from single images; Genie 3 generates them from text prompts in real time at 24fps with far greater diversity and consistency. |
| [Sora vs Genie 3](https://world-models.io/en/compare/sora-vs-genie-3/) | Sora vs Genie 3 | Two leading generative world models with opposite design goals: Sora targets long, cinematic, non-interactive clips from text, while Genie 3 trades fidelity for real-time controllable worlds you can actually play in. |
| [Pandora vs Genie 2](https://world-models.io/en/compare/pandora-vs-genie-2/) | Pandora vs Genie 2 | Both generate explorable 3D-feeling environments, but with different control surfaces: Pandora accepts free-form text actions through an LLM backbone, while Genie 2 conditions on a single seed image and learned latent actions. |
| [Genie 3 vs NVIDIA Cosmos](https://world-models.io/en/compare/genie-3-vs-nvidia-cosmos/) | Genie 3 vs NVIDIA Cosmos | Two green-index frontier systems with different ambitions. Genie 3 is a real-time text-to-world interactive generator, while NVIDIA Cosmos is a broad physical-AI platform optimized for simulation infrastructure, robotics, and industrial world modeling. |
| [Genie 3 vs V-JEPA 2](https://world-models.io/en/compare/genie-3-vs-v-jepa-2/) | Genie 3 vs V-JEPA 2 | Two green-index leaders that represent different frontier philosophies. Genie 3 is an interactive generative world model that turns text into playable environments, while V-JEPA 2 is a self-supervised latent predictor optimized for physical reasoning and zero-shot robot planning. |
| [NVIDIA Cosmos vs V-JEPA 2](https://world-models.io/en/compare/nvidia-cosmos-vs-v-jepa-2/) | NVIDIA Cosmos vs V-JEPA 2 | Two green-index foundation-scale leaders with different views of world modeling. Cosmos emphasizes a platform for physical-AI simulation and generation, while V-JEPA 2 emphasizes self-supervised predictive representations for visual understanding and robot control. |
| [PlayWorld vs V-JEPA 2](https://world-models.io/en/compare/playworld-vs-v-jepa-2/) | PlayWorld vs V-JEPA 2 | Two green-index models pushing robotics-relevant world understanding in different ways. PlayWorld is a robot-play simulator for manipulation and policy improvement, while V-JEPA 2 is a self-supervised video predictor optimized for physical reasoning and zero-shot robot planning. |
| [DreamerV3 vs PlayWorld](https://world-models.io/en/compare/dreamer-v3-vs-playworld/) | DreamerV3 vs PlayWorld | Two green-index leaders for acting under learned dynamics, but with different centers of gravity. DreamerV3 is the canonical imagination-based general RL agent, while PlayWorld is a manipulation-centric robot simulator trained from autonomous play data. |

## Research Topics Referencing This Model

Connected research areas surfaced directly in static HTML.

| Topic | Summary |
| --- | --- |
| [Video World Models](https://world-models.io/en/research/video-world-models/) | How video world models and video generation transformers learn physics, temporal consistency, and interactive simulation from large-scale video, from OpenAI Sora and Google DeepMind Genie to NVIDIA Cosmos and V-JEPA. |
| [Diffusion World Models](https://world-models.io/en/research/diffusion-world-models/) | How diffusion world models generate future states, preserve richer visual detail, and power video simulation systems such as DIAMOND, Sora, and Cosmos. |
| [Language-Conditioned World Models](https://world-models.io/en/research/language-conditioned-world-models/) | How language-conditioned world models use text prompts or natural-language actions to control simulation, planning, and embodied behavior across Pandora, 3D-VLA, RT-2, and hybrid systems. |

## Timeline Mentions

Recent timeline events connected to this model.

| Event | Published | Source | Summary |
| --- | --- | --- | --- |
| [Google DeepMind unveils Genie 3: real-time interactive world model at 720p/24fps](https://world-models.io/en/timeline/deepmind-genie-3-release/) | 2026-04-22 | Google DeepMind | DeepMind announces Genie 3, a real-time foundation world model generating interactive 3D environments at 720p and 24 fps from text prompts. |

## Frequently Asked Questions

FAQ answers rendered directly into static HTML for extractable responses.

### What's new in Genie 3 compared to Genie 2?

Genie 3 introduces real-time generation at 24fps (vs offline in Genie 2), supports text prompts instead of just image conditioning, generates far more diverse environments, and is publicly accessible via Project Genie.

### Can I try Genie 3?

Yes: Google AI Ultra subscribers in the U.S. can access Project Genie, a research prototype built on Genie 3, to create and explore AI-generated worlds.

## Quick Answer

Short extractable summary preserved directly in static HTML.

- Genie 3 is a real-time interactive 3d environment generation developed by Google DeepMind in 2025 for 3d world generation.
- Use this page when you need a fast read on how Genie 3 fits into the generative world model landscape, then validate the details in the benchmarks, citations, and related pages.
- A key strength surfaced in the editorial record is first real-time interactive world model.

## Editorial Trust Signals

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: 2026-04-10.

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.](https://world-models.io/en/editorial-policy/)

Primary sources only

Last reviewed date visible

Methodology documented

Source links included

Correction channel published

Review date visible

## External Sources

Primary model and lab sources embedded in static HTML.

- [Genie 3 paper](https://deepmind.google/blog/genie-3-a-new-frontier-for-world-models/) Google DeepMind primary research source
- [Genie 3 official page](https://deepmind.google/models/genie/) Google DeepMind official page
- [Genie 3 cited source](https://arxiv.org/abs/1803.10122) Parker-Holder & Fruchter, 2025. Genie 3: A New Frontier for World Models. Google DeepMind Blog.
- [Google DeepMind official website](https://deepmind.google) Related lab

## References

1. [[1] Parker-Holder & Fruchter, 2025. Genie 3: A New Frontier for World Models. Google DeepMind Blog.](https://arxiv.org/abs/1803.10122)