🍀
Highlights
- Pro
Starred repositories
[Tech Report] Context Scaling: Scaling Properties of Text Conditioning in Visual Generation
A Call of Duty-quality FPS in Three.js, built from a single prompt.
Infinite Worlds with Versatile Interactions
EdgeBench: Unveiling scaling laws of learning from real-world environments
Boogu-Image-0.1 is an Apache-2.0 open-source image generation and editing model family that delivers near-closed-source performance with an order of magnitude less data.
A CLI tool to query your local WeChat data — chat history, contacts, sessions, favorites, and more. Designed for LLM integration.
ViGeo: Towards Consistent Video Geometry Estimation
Official implementation of R^3: 3D Reconstruction via Relative Regression.
Stable and Efficient Reinforcement Learning for Trillion-Parameter LLMs
[ICLR 2026] PyTorch implementation of "The Less You Depend, The More You Learn: Synthesizing Novel Views from Sparse, Unposed Images with Mimimal 3D Knowledge".
Agentic World Modeling: Foundations, Capabilities, Laws, and Beyond
[SIGGRAPH Asia 2026] AnyRecon: Arbitrary-View 3D Reconstruction with Video Diffusion Model
Auto-loop execution workflow with quality gates for Claude Code. Automatically decomposes tasks, implements code, runs tests, and iterates through quality gates until completion.
A feed-forward 3D foundation model for reconstructing scenes from streaming data
Unified Codebase for Advanced World Models.
[CVPR 2026 (Highlight)] Scal3R: Scalable Test-Time Training for Large-Scale 3D Reconstruction
🎭 Know yourself as a developer. One command → AI analyzes your coding history → beautiful personality portrait + persona skill. Works with Claude Code & Codex.
Bridge local AI coding agents (Claude Code, Cursor, Gemini CLI, Codex) to messaging platforms (Feishu/Lark, DingTalk, Slack, Telegram, Discord, LINE, WeChat Work). Chat with your AI dev assistant f…
Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models
Official code for "LagerNVS Latent Geometry for Fully Neural Real-time Novel View Synthesis" (CVPR 2026)