Stars
Show usage stats for OpenAI Codex and Claude Code, without having to login.
Toolkit for linearizing PDFs for LLM datasets/training
AutoGaze automatically removes redundant patches in a video, reducing #tokens in ViT/MLLM by 4x-100x.
Transforms complex documents like PDFs into LLM-ready markdown/JSON for your Agentic workflows.
PDF Parser for AI-ready data. Automate PDF accessibility. Open-source.
For people who get distracted by agents. A native Rust/GPUI control plane for running Claude Code, Codex, Cursor, and OpenCode side by side — because if you're going to be squirrely, you might as w…
Official implementation for paper: Learning Self-Correction in Vision–Language Models via Rollout Augmentation
Teams-first Multi-agent orchestration for Claude Code
Run OpenClaw more securely inside NVIDIA OpenShell with managed inference
SparkVSR: Interactive Video Super-Resolution via Sparse Keyframe Propagation
Official implementation of "Planning in 8 Tokens: A Compact Discrete Tokenizer for Latent World Model" (CVPR 2026)
Web interface for browsing, search and filtering recent arxiv submissions
Youtu-VL: Unleashing Visual Potential via Unified Vision-Language Supervision
OpenSeeker: A search agent with open-source data and models
Official Implementation of "Visual-ERM: Reward Modeling for Visual Equivalence"
Seoul World Model: Grounding World Simulation Models in a Real-World Metropolis
Next-generation Albumentations: dual-licensed for open-source and commercial use
Code for Agentic Very Long Video Understanding (EGAgent)
A coding agent with web search in a single Python script, around 100 lines for educational purposes.
A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs dir…
Official implementation of "LittleBit: Ultra Low-Bit Quantization via Latent Factorization" (NeurIPS 2025)
Telegram ↔ tmux bridge for Claude Code: 1 topic = 1 window = 1 session