Lists (2)
Starred repositories
MAGI-2-preview: Scaling Video Generation Models Efficiently
"DeepCode: Open Agentic Coding (Paper2Code & Text2Web & Text2Backend)"
OpenCE (Open Context Engineering): A community toolkit to implement, evaluate, and combine LLM context strategies (RAG, ACE, Compression). Evolved from the `ACE-open` reproduction.
[Official Repo] JoyAI-Video-Edit: Real-Time Open-Ended Video Editing with Autoregressive Diffusion
A GPU with transformation & lighting, rasterization, texture mapping, flat shading, double buffer, z-buffer
Official inference code for SoulX-Singer: Towards High-Quality Zero-Shot Singing Voice Synthesis
DeepSeek-V4-Flash-0731 on 4x CMP 170HX (sm_80): 98 tok/s decode, ~5300 tok/s prefill. DSpark speculative decoding under pipeline parallelism, which vLLM does not support upstream.
Invidious is an alternative front-end to YouTube
TencentDB Agent Memory is a team-level memory hub for AI Agents — turning conversations, docs, and code into four reusable memory assets (Chat Memory, Skill, LLM-Wiki, Code-Graph) that are governed…
- 100+ skills across technology, business, marketing, creative - Skill normalization with aliases and transferability scores - Proficiency markers and typical roles - CC BY 4.0 licensed for attribu…
new skills taxonomy using TextKernel data
AI-native extension for SQLite that brings on-device inference, embedding generation, and model interaction directly into your database.
Run the full 2.78-trillion-parameter Kimi K3 model beyond available RAM by streaming activated weights directly from NVMe. A dependency-free, embeddable C inference engine.
The iOS and MacOS port of Xenia, the Xbox 360 Emulator Research project.
Game engine for 2D and 3D projects with entity component system (ECS) and data-oriented design
An experimental Gecko-based web browser for iOS 13+.
Gemma 4 26B-A4B inference in ~2 GB of RAM on any M-series MacBook
A realtime voice runtime that keeps Agents talking, working, and present. Real-time Voice Runtime for AI Agents
12 Weeks, 24 Lessons, AI for All!
CMP 170HX technical wiki: silicon, firmware, the community unlock, procedures and open problems. 55 pages.
A typeface that protects written content by poisoning unauthorized AI training datasets.
Code release for 360monodepth. With our framework we achieve monocular depth estimation for high resolution 360° images based on aligning and blending perspective depth maps.
DeepSeek V4 Flash DSpark 1M NVFP4 KV recipe for 2x DGX Spark
From a single casual image to a visually consistent and physically stable interactive 3D scene.