New: the Timeline is live. Track world model releases, papers, and benchmark updates in real time.
world-models.io
The Knowledge Hub for AI World Models

Genie 3 vs Genie 2

Genie 3 vs Genie 2 compares two generations of DeepMind's interactive world model: Genie 2 generates 3D environments from images; Genie 3 generates them from text prompts in real time at 24fps.

robotics model-based-rl simulation embodied-ai

Comparison Overview

Main comparison summary preserved directly in static HTML.

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.

Verdict

Primary editorial conclusion preserved for non-JS crawlers and readers.

Genie 3 represents a generational leap: real-time generation, text conditioning, and public access via Project Genie. Genie 2 was the research proof-of-concept; Genie 3 is the productized world model. Genie 3 remains limited to several minutes of continuous interaction, with visual memory reaching back about one minute.

Key Differences

Extractable difference list generated from the comparison table.

  • Input: Genie 3 - Text prompts; Genie 2 - Single image.
  • Generation Speed: Genie 3 - Real-time at 24fps; Genie 2 - Offline generation.
  • Resolution: Genie 3 - 720p; Genie 2 - Variable.
  • Consistency Duration: Genie 3 - ~1 minute; Genie 2 - ~10-20 seconds.
  • Environment Diversity: Genie 3 - Unprecedented variety; Genie 2 - Limited to image-conditioned.

When To Use Each

Static decision guidance for no-JS readers.

Choose Genie 3 when...

Choose Genie 3 when its capabilities best match your research or deployment requirements.

Choose Genie 2 when...

Choose Genie 2 when its capabilities best match your research or deployment requirements.

Comparison Table

Genie 3 represents a generational leap: real-time generation, text conditioning, and public access via Project Genie. Genie 2 was the research proof-of-concept; Genie 3 is the productized world model. Genie 3 remains limited to several minutes of continuous interaction, with visual memory reaching back about one minute.

DimensionGenie 3Genie 2
InputText promptsSingle image
Generation SpeedReal-time at 24fpsOffline generation
Resolution720pVariable
Consistency Duration~1 minute~10-20 seconds
Environment DiversityUnprecedented varietyLimited to image-conditioned
Public AccessYes (Project Genie)Research only
Year20252024

Performance Index Snapshot

High-level scoring context for the models referenced in this comparison.

ModelCategoryIndex v1.1Confidence
Genie 3Generative World Model89/100medium
Genie 2Generative World Model79/100medium
GenieGenerative World Model57/100medium
OASISGenerative World Model66/100medium

Frequently Asked Questions

FAQ answers rendered directly into static HTML for extractable responses.

Can I try Genie 3?

Yes: Google AI Ultra subscribers in the U.S. can access Project Genie to create and explore AI-generated worlds.

Is Genie 3 open-source?

No, Genie 3 is closed-source. It is only available through the Project Genie product.

Quick Answer

Short extractable summary preserved directly in static HTML.

  • Genie 3 vs Genie 2: this page compares where each system is stronger instead of forcing a universal winner.
  • Use the verdict for the short answer, then validate the trade-offs in the table, evidence sources, and benchmark context.
  • Related models and source links help connect this comparison to the broader world models landscape.

Editorial Trust Signals

Editorial provenance and refresh policy preserved directly in static HTML.

Published by world-models.io editorial board.

Lead editor Bernard Grenat.

This comparison page publishes a direct answer, explicit trade-offs, and source-backed evidence that can be validated against primary materials.

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: 2026-06-21.

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 sources onlyLast reviewed date visibleMethodology documentedSource links included

External Sources

Primary papers and official sources for the models discussed on this comparison page.