📹 A more flexible framework that can generate videos at any resolution and creates videos from images.

minimind-v-RL 基于minimind-v多模态预训练模型进行GRPO强化学习训练，以实现小模型思考能力. Minimind-v-RL conducts GRPO reinforcement learning training based on minimind-v multimodal pre training model to achieve small model think…

谷歌新书Agent设计模式(agentic design patterns)最佳中文版，持续优化。附：在线阅读、pdf和epub电子书下载。

A theoretical reconstruction of the Claude Mythos architecture, built from first principles using the available research literature.

MiniCPM4 & MiniCPM4.1: Ultra-Efficient LLMs on End Devices, achieving 3+ generation speedup on reasoning tasks

text and image to video generation: CogVideoX (2024) and CogVideo (ICLR 2023)

Implementation of Denoising Diffusion Probabilistic Model in Pytorch

Thinking with Videos from Open-Source Priors. We reproduce chain-of-frames visual reasoning by fine-tuning open-source video models. Give it a star 🌟 if you find it useful.

🎓从0开始训练一个大模型Minimind项目的超详细解析，包括但不限于用到的架构，算法，以及大模型面试经验

PCSim: LiDAR Point Cloud Simulation and Sensor Placement! Code of [ICRA 2023] "Analyzing Infrastructure LiDAR Placement with Realistic LiDAR Simulation Library" and [ICCV 2023] "Optimizing the Plac…

Project Lyra: Open Generative 3D World Models

[NeurIPS 2025] 𝒳-Scene: Large-Scale Driving Scene Generation with High Fidelity and Flexible Controllability

[CVPR 2025] UniScene: Unified Occupancy-centric Driving Scene Generation

[CVPR 2026] The official PyTorch implementation of the "Vision Transformer Needs More Than Registers".

Python code for "Probabilistic Machine learning" book by Kevin Murphy

Graphic notes on Gilbert Strang's "Linear Algebra for Everyone", 线性代数的艺术中文版, 欢迎PR.

Enjoy the magic of Diffusion models!

[ICLR 2026] FantasyWorld: Geometry-Consistent World Modeling via Unified Video and 3D Prediction

Code to pretrain, fine-tune, and evaluate DreamZero and run sim & real-world evals

A PyTorch port of DeepMind's Disco103 — the meta-learned reinforcement learning update rule from Discovering State-of-the-art Reinforcement Learning Algorithms (Nature, 2025).

Accompanying code for "Discovering State-of-the-art Reinforcement Algorithms" Nature publication

[ICLR 2026] ReCogDrive: A Reinforced Cognitive Framework for End-to-End Autonomous Driving

🤗 LeRobot: Making AI for Robotics more accessible with end-to-end learning

StarVLA: A Lego-like Codebase for Vision-Language-Action Model Developing

本项目旨在分享大模型相关技术原理以及实战经验（大模型工程化、大模型应用落地）