GWM (Gaussian World Models)
https://gaussian-world-model.github.io/
Encodes scene into 3D Gaussian Splat representations via 3D VAE; latent Diffusion Transformer predicts future Gaussian Splat states conditioned on robot actions. 3D scene representation persists across time steps as structural scene memory rather than flat features.
At a glance
- Type
- 3D Gaussian Splatting latent world model
- Tier
- T3
- Created
- 2025-08-25 (GWM: Towards Scalable Gaussian World Models for Robotic Manipulation; arxiv 2508.17600 submitted Aug 25…
- Latest release
- not applicable — not OSS
- License
- not applicable — not OSS
- GitHub
- not applicable — no GitHub repo
- Pricing
- searched not found
- Funding
- searched not found
Taxonomy
- storage
- parametric
- retrieval
- parametric-recall
- persistence
- session
- update
- overwrite
- unit
- scene-graph
- governance
- inspectable
- conflict
- overwrite
When to use
Optimised for: real-time perception + spatial reasoning + multi-modal sensor fusion
Anti-fit: not for non-embodied / non-physical use cases
Pros & cons
Pros
Efficient world-model representation using Gaussian distributions over states.
Cons
Research-stage; expressiveness limited by Gaussian assumption.
Claims & capabilities
ICCV 2025. Action-conditioned 3D video prediction; visual representation learning for imitation; neural simulator for model-based RL.
Technical surface
- API surface
- not applicable — research paper
- Backend storage
- not applicable — research paper
- Deployment
- searched not found
- Embedding model
- not applicable — research paper
- Multi-tenancy
- not applicable — research paper
- MCP
- not applicable — research paper, no deployed product
- A2A
- not applicable — research paper, no deployed product
- OpenTelemetry
- not applicable — research paper, no deployed product
Similar systems
Other vertical / domain-specific ai memory in the catalog, ranked by inbound references.
- NVIDIA ReMEmbR T3
Builds long-horizon memory by captioning video segments with VILA, storing captions with timestamps + 3D position coordinates in MilvusDB. At query time, LLM iterates retrieval across text, time, and position modalities. Deployed on Nova Carter robot (Jetson Orin).
- Abridge T1
Clinician-assist ambient documentation. Source mapping: every AI-generated summary element traced back to the source utterance. Audit-and-trust layer over episodic memory. Built on proprietary 1.5M+ medical-encounter dataset.
- ASAPP GenerativeAgent T1
Treats memory as first-class architecture. Captures the digital footprint of every interaction; retrieves preference and history at engagement time. Public example: airline knowing a frequent flyer wants aisle seats with her son — preference-aware, not just history-lookup.
- BenevolentAI T1
Target identification / drug repurposing / mechanism tracing. 85+ data sources, petabyte-scale, rebuilt every few weeks. Wet-lab results re-enter the graph and shift downstream predictions — institutional experimental memory closing a feedback loop.
- Causaly T1
Drug discovery / target identification / causal mechanism tracing. The graph is the memory: 7 years of curated biomedical cause-effect relationships compounding with each new ingestion. Scientific RAG retrieves from a versioned causal substrate.
- Character.ai T1
Chat Memories (user-defined facts), auto-memories for c.ai+ subscribers, pinned memories, in-context retention. PipSqueak 2 model (April 2026) reduces in-conversation drift. Memory Visualization meter forthcoming.