--- title: "IRIS vs DreamerV3 | Comparison | world-models.io" canonical: "https://world-models.io/en/compare/iris-vs-dreamer-v3/" description: "Comparing IRIS (autoregressive Transformer world model) and DreamerV3 (RSSM) for model-based reinforcement learning." author: "Bernard Grenat" generated: "2026-10-23" --- New: the Timeline is live. Track world model releases, papers, and benchmark updates in real time. The Knowledge Hub for AI World Models # IRIS vs DreamerV3 IRIS and DreamerV3 are both leading model-based RL agents but use fundamentally different world model architectures: autoregressive token prediction vs. RSSM latent dynamics. 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/). ## Comparison Overview Main comparison summary preserved directly in static HTML. IRIS and DreamerV3 are both leading model-based RL agents but use fundamentally different world model architectures: autoregressive token prediction vs. RSSM latent dynamics. ## Verdict Primary editorial conclusion preserved for non-JS crawlers and readers. DreamerV3 is the stronger performer and more general-purpose agent. IRIS is conceptually important because it bridges world modeling and language modeling, suggesting that advances in LLM architectures could directly benefit world models. As Transformers continue to scale, IRIS-style approaches may narrow the gap. ## Key Differences Extractable difference list generated from the comparison table. - **World Model Type:** IRIS - Autoregressive (next-token prediction); DreamerV3 - RSSM (latent dynamics). - **Representation:** IRIS - VQ-VAE discrete tokens; DreamerV3 - Discrete categorical latents. - **Dynamics Model:** IRIS - Transformer; DreamerV3 - GRU-based recurrent model. - **Connection To:** IRIS - Language modeling (GPT-style); DreamerV3 - State-space models. - **Atari 100K:** IRIS - 1.046x human; DreamerV3 - 2.01x human (SOTA). ## When To Use Each Static decision guidance for no-JS readers. ### Choose IRIS when... Choose IRIS when its documented capabilities match your research or deployment requirements. ### Choose DreamerV3 when... Choose DreamerV3 when its documented capabilities match your research or deployment requirements. ## Comparison Table DreamerV3 is the stronger performer and more general-purpose agent. IRIS is conceptually important because it bridges world modeling and language modeling, suggesting that advances in LLM architectures could directly benefit world models. As Transformers continue to scale, IRIS-style approaches may narrow the gap. | Dimension | IRIS | DreamerV3 | | --- | --- | --- | | World Model Type | Autoregressive (next-token prediction) | RSSM (latent dynamics) | | Representation | VQ-VAE discrete tokens | Discrete categorical latents | | Dynamics Model | Transformer | GRU-based recurrent model | | Connection To | Language modeling (GPT-style) | State-space models | | Atari 100K | 1.046x human | 2.01x human (SOTA) | | Domain Breadth | Atari (primary) | Atari, DMControl, Minecraft, etc. | | Scalability Path | Scales with Transformer advances | Scales with RSSM improvements | ## Performance Index Snapshot High-level scoring context for the models referenced in this comparison. | Model | Category | Index v1.1 | Confidence | | --- | --- | --- | --- | | IRIS | Model-Based RL | 65/100 | medium | | DreamerV3 | Model-Based RL | 88/100 | high | | DIAMOND | Model-Based RL | 64/100 | medium | ## Frequently Asked Questions FAQ answers rendered directly into static HTML for extractable responses. ### Will Transformer-based world models surpass RSSM? Possibly. As Transformers scale and become more efficient, autoregressive world models like IRIS could benefit from the same innovations driving LLM progress. The RSSM's advantage lies in its inductive bias for temporal dynamics. ## Quick Answer Short extractable summary preserved directly in static HTML. - IRIS vs DreamerV3: 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: 2026-09-27. 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 papers and official sources for the models discussed on this comparison page. - [IRIS paper](https://arxiv.org/abs/2209.00588) Microsoft Research primary research source - [IRIS cited source](https://arxiv.org/abs/1803.10122) Micheli et al., 2023. Transformers are Sample-Efficient World Models. ICLR 2023. - [DreamerV3 paper](https://arxiv.org/abs/2301.04104) Google DeepMind primary research source - [DIAMOND paper](https://arxiv.org/abs/2405.12399) Microsoft Research / University of Geneva primary research source