A feed-forward 3D foundation model for reconstructing scenes from streaming data

Official code base for LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels

selfOS — your personal operating system. Compiles notes, AI conversations, and documents into a persistent knowledge graph. Zero-friction capture, oral history interviews, hybrid search.

LoMa: Local Feature Matching Revisited

Allen Institute for AI: WildDet3D: Scaling Promptable 3D Detection in the Wild

Code for the Molmo2 Vision-Language Model

Autoresearch for GPU kernels. Give it any PyTorch model, go to sleep, wake up to optimized Triton kernels.

🏛️ 三省六部制 · OpenClaw Multi-Agent Orchestration System — 9 specialized AI agents with real-time dashboard, model config, and full audit trails

Block-Recurrent Dynamics in ViTs 🦖

Fast and memory-efficient exact attention

Claude Code VS Code extension patched for Force Local mode — run CLI locally, proxy file ops to remote server via VS Code Remote SSH

Uses SAM3 to transfer 2D semantics onto 3D point clouds, producing segmented training data and efficient scene graphs for indoor environments.

An open source library designed to provide community examples of Joint Embedding Predictive Architectures (JEPAs). It contains code and examples for learning representations from images, video, and…

Fast symbolic computation, code generation, and nonlinear optimization for robotics

Visual Causal Flow

Causal video-action world model for generalist robot control

Gain deeper insights into your favourite open-source GitHub repositories. Explore star trends showing daily stats.

Advancing Open-source World Models

A Pragmatic VLA Foundation Model

Masked Depth Modeling for Spatial Perception

Claude Code SDK Demos

[TPAMI 2026] Interacted Planes Reveal 3D Line Mapping

Pixio: a capable vision encoder dedicated to dense prediction, simply by pixel reconstruction

Headless remote desktop setup for NVIDIA DGX SPARK using Sunshine streaming

[NeurIPS 2025] Official code for JAFAR: Jack up Any Feature at Any Resolution

[NeurIPS 2025] the official project page of a paper, "PLANA3R: Zero-shot Metric Planar 3D Reconstruction via Feed-Forward Planar Splatting"

The simplest, fastest repository for training/finetuning medium-sized GPTs.

The Open-Source Multimodal AI Agent Stack: Connecting Cutting-Edge AI Models and Agent Infra