TurboVLA: Real-Time Vision-Language-Action Model at 32 Hz on an RTX 4090 with <1 GB VRAM

Open Frontier Intelligence

DBL-Diffusion: Explicit Layer Modeling for Video Object Insertion and Layer Decomposition

A straightforward method for training your LLM, from downloading data to generating text.

Unlimited OCR Works: Welcome the Era of One-shot Long-horizon Parsing.

Large Language-and-Vision Assistant for Lunar Exploration (LLaVA-LE)

Large-Scale Multimodal Dataset of Astronomical Data

Code for the CVPR 2026 AssemblyBench paper

[AAAI 2026] Official Implementation of NURBGen: High-Fidelity Text-to-CAD Generation through LLM-Driven NURBS Modeling

[ICLR 2025] Official PyTorch Implementation of Gated Delta Networks: Improving Mamba2 with Delta Rule

NVIDIA Cosmos is an open platform of world models, datasets, and tools that enables developers to build Physical AI for robots, autonomous vehicles, smart infrastructure, and more.

This repository is a custom node in ComfyUI. This is a program that allows you to use Huggingface Diffusers module with ComfyUI. Additionally, Stream Diffusion is also available.

Long Korean Document을 위한 VLM 벤치마크

Official implementation of AsymFlow, pi-Flow, GMFlow

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.

한국인을 위한 스킬 모음집 - 에이전트를 한국인으로

OpenVLA: An open-source vision-language-action model for robotic manipulation.

[CVPR 2026 Highlight] EffectErase: Joint Video Object Removal and Insertion for High-Quality Effect Erasing

Seoul World Model: Grounding World Simulation Models in a Real-World Metropolis

[WACV 2026] Official implementation of "Edge-Aware Image Manipulation via Diffusion Models with a Novel Structure-Preservation Loss"

[CVPR2024 Highlight] VBench - We Evaluate Video Generation

[CVPR 2026 Highlight🔥] MotionCrafter: Dense Geometry and Motion Reconstruction with a 4D VAE

General plug-and-play inference library for Recursive Language Models (RLMs), supporting various sandboxes.

nanoRLHF: from-scratch journey into how LLMs and RLHF really work.

Official repository for K-EXAONE built by LG AI Research

self-play 방법론중, 모델의 behavior를 고려하여 diversity를 추가는 방법을 논의하는 git-hub입니다.