Official implementation of paper "Unified World Models: Memory-Augmented Planning and Foresight for Visual Navigation"

A blueprint for building production-ready RAG systems that minimize hallucination, featuring switchable 3-step (Speed) and 4-step (Precision) pipelines.

minerproxy矿池中转，原创minerproxy，唯一正版，性能强大, 功能齐全, 9000台无压力不崩溃，BTC ETC ETH LTC等全币种无损抽水，体验拉满的minerproxy, 软防CC, 动态难度调整抽水，不爆内存。支持绝大部分币种的转发、加密、自定义抽水、精确到单台设备的24小时数据统计、自定义隧道推送工具、修改矿池本地算力

Hierarchical Reasoning Model Official Release

Align Anything: Training All-modality Model with Feedback

"AutoAgent: Fully-Automated and Zero-Code LLM Agent Framework"

ANIMAT is the first AI platform to integrate MMD and facial tracking for dynamic 3D Model, enabling realistic customization and upgrade of AgentAI.

Resources of our paper "FilmAgent: A Multi-Agent Framework for End-to-End Film Automation in Virtual 3D Spaces". New versions in the making!

PVPAI LLM 🔥The First Open-Source DeFAI Large Language Model Powered by DeepSeek.

"LightRAG: Simple and Fast Retrieval-Augmented Generation"

Champ: Controllable and Consistent Human Image Animation with 3D Parametric Guidance

[ICLR 2024] Official implementation of "TimeMixer: Decomposable Multiscale Mixing for Time Series Forecasting"

The next generation deep reinforcement learning tookit

vue3常用的hooks集合

RRT based path planning