JeremyBYU / polylidar

Official code for article "LLMLight: Large Language Models as Traffic Signal Control Agents".

Safe Multi-Agent Isaac Gym benchmark for safe multi-agent reinforcement learning research.

GIM: Learning Generalizable Image Matcher From Internet Videos (ICLR 2024 Spotlight)

[ICML 2024] 3D-VLA: A 3D Vision-Language-Action Generative World Model

A humanoid bipedal walking control repo using NMPC and WBC, and simulation with mujoco. Contact: xuejl2001@mail.ustc.edu.cn

[ECCV2024] API code for T-Rex2: Towards Generic Object Detection via Text-Visual Prompt Synergy

Original reference implementation of "3D Gaussian Splatting for Real-Time Radiance Field Rendering"

Powerful Light Artificial NEuRon inference framework for CNN

Segment Anything in High Quality [NeurIPS 2023]

Warp 是一个用于编写高性能仿真和图形代码的 Python 框架

fastllm是后端无依赖的高性能大模型推理库。同时支持张量并行推理稠密模型和混合模式推理MOE模型，任意10G以上显卡即可推理满血DeepSeek。双路9004/9005服务器+单显卡部署DeepSeek满血满精度原版模型，单并发20tps；INT4量化模型单并发30tps，多并发可达60+。

#计算机科学#A hyperparameter optimization framework

Visual SLAM/odometry package based on NVIDIA-accelerated cuVSLAM

#计算机科学#MuJoCo fruit fly body model and locomotion RL tasks

3D高斯论文，持续更新，欢迎交流讨论。