S-Space: Exploring Spatial Workspace in Multimodal Models

Local UI to run and train LLMs and diffusion models. Supports GGUF, MLX, Qwen3.8, DeepSeek-V4, MiniMax-H3, Gemma 4, FLUX and more.

MiniCPM5: SOTA on-device LLMs, small yet powerful.

PyTorch implementation of JiT https://arxiv.org/abs/2511.13720

[ICLR 2026] SpaceControl: Introducing Test-Time Spatial Control to 3D Generative Modeling.

[SIGGRAPH Asia 26]PAct: Part-Decomposed Single-View Articulated Object Generation

Research and development (R&D) is crucial for the enhancement of industrial productivity, especially in the AI era, where the core aspects of R&D are mainly focused on data and models. We are commi…

A Foundation Model for Generalist Gaming Agents

Benchmarking MLLMs for Parametric 3D Generation and Structural Reasoning (Text-to-3D, Image-to-3D, Assembly-3D)

Agentic 3D Modeling with Procedural Control — turns a text prompt into an editable parametric program, with optional per-part materials and articulation.

[CVPR 2026] Stereo World Model

Official training code for TripoSplat: Generative 3D Gaussians with Learned Density Control [SIGGRAPH 2026]

TripoSplat converts a single 2D image into high-quality and variable number of 3D Gaussians, developed by TripoAI.

Simulation platform for general-purpose robotics & embodied AI learning.

Code as World: Agentic Discovery of Executable World Representations for Physical Reasoning

Code World Model Official Repo

Qwen3.8-Flash-Next is the foundation model developed by Qwen Team, Alibaba Group.

[3DV 2025] Official Implementation of paper "PIR: Photometric Inverse Rendering with Shading Cues Modeling and Surface Reflectance Regularization"

Official code for paper "SCULPT: Subtractive Composition for 3D Part Generation"

Code repository for Qwen-3D: A Generalist 3D Vision-Language Model for Spatial Understanding

An curated list for feed-forward 3D scene modeling, including research directions, datasets, and applications.

[TMM 2026] FAME: Mask-Guided Multimodal Fashion Attribute Editing

Codebase for "Beyond Pixel Histories: World Models with Persistent 3D State" (ICML 26)

[ICLR 2026] PartSAM: A Scalable Promptable Part Segmentation Model Trained on Native 3D Data

[NeurIPS 2025] SpatialLM: Training Large Language Models for Structured Indoor Modeling