embodied-ai

DORA (Dataflow-Oriented Robotic Architecture) is middleware designed to streamline and simplify the creation of AI-based robotic applications. It offers low latency, composable, and distributed dataflow capabilities. Applications are modeled as directed graphs, also referred to as pipelines.

Open-source AI assistant ecosystem with MCP integrations, multimodal workflows, IoT support, and cross-platform voice interaction.

🦦 Otter, a multi-modal model based on OpenFlamingo (open-sourced version of DeepMind's Flamingo), trained on MIMIC-IT and showcasing improved instruction-following and in-context learning ability.

Awesome things about LLM-powered agents. Papers / Repos / Blogs / ...

UnrealCV: Connecting Computer Vision to Unreal Engine

A library for differentiable nonlinear optimization

A comprehensive list of papers for the definition of World Models and using World Models for General Video Generation, Embodied AI, and Autonomous Driving, including papers, codes, and related websites.

[TMLR 2025🔥] A survey for the autoregressive models in vision.

RAI is a vendor agnostic agentic framework for Physical AI robotics, utilizing ROS 2 tools to perform complex actions, defined scenarios, free interface execution, log summaries, voice interaction and more.

[CVPR 2025] RoboBrain: A Unified Brain Model for Robotic Manipulation from Abstract to Concrete. Official Repository.

《多模态大模型：新一代人工智能技术范式》配套教学资源

Official Code for "Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents"

A collection of vision-language-action model post-training methods.

a comprehensive and critical synthesis of the emerging role of GenAI across the full autonomous driving stack

"World Models in a Closed-Loop World" (ICLR'26 Oral)

[ICCV 2025] Perspective-Invariant 3D Object Detection

DataClaw: Agentic Tailoring Multimodal Data from Raw Streams — coming soon (code, weights, dataset & DataClaw-val upon acceptance).

[NeurIPS 2024] GenRL: Multimodal-foundation world models enable grounding language and video prompts into embodied domains, by turning them into sequences of latent world model states. Latent state sequences can be decoded using the decoder of the model, allowing visualization of the expected behavior, before training the agent to execute it.

[ICML 2024] The offical Implementation of "DecisionNCE: Embodied Multimodal Representations via Implicit Preference Learning"

An benchmark for evaluating the capabilities of large vision-language models (LVLMs)