OmniSpace

本地优先的全栈 AI 创作工作站:AI对话 · AI漫画 · 漫剧创作 · 写作台 · 知识学习五大模块;vLLM / llama.cpp / Transformers 多引擎本地推理,显存智能调度、舱壁隔离自愈

What is OmniSpace?

本地优先的全栈 AI 创作工作站:AI对话 · AI漫画 · 漫剧创作 · 写作台 · 知识学习五大模块;vLLM / llama.cpp / Transformers 多引擎本地推理,显存智能调度、舱壁隔离自愈、仅供学习使用,欢迎交流指正。(最新进度(除插件系统)受到协议限制,暂时无法提交到仓库)

Overview of OmniSpace

What it is

OmniSpace is a local-first, full-stack AI creative workstation that integrates five modules: AI dialogue, AI comic creation, AI manga (animated comic) creation, a writing desk, and knowledge learning. It supports local inference on consumer GPUs using multiple engines (vLLM, llama.cpp, Transformers) and optionally connects to cloud APIs with user-provided keys. The project is open source and intended for learning purposes only.

Key features

  • Local-first inference: All model inference runs on the user's own GPU by default; data never leaves the machine. Cloud API is an optional, clearly labeled setting.
  • VRAM scheduler: A self-developed coordinator that manages VRAM allocation, enabling 16GB consumer GPUs to run various creative pipelines with automatic loading/unloading and mutual exclusion.
  • Bulkhead isolation and self-healing: Each major module has an independent health ledger; continuous anomalies trigger a recovery ladder (engine restart → verification → recovery guide + safe mode).
  • Data lifecycle management: Deletion with three-level confirmation, storage cleanup center, backup encryption (AES-GCM), daily automatic backup, power-off self-recovery, operation audit logs.
  • Plugin system (CuteMamen): Subprocess sandbox isolation, AST static security check, three-tier trust badges, hot-reload without restart.
  • Multi-engine hot-switching: vLLM, llama.cpp (GGUF), and Transformers backends can be switched at runtime.
  • Comprehensive testing: 1132 pytest cases (with explicit GPU toggle), 101 vitest cases, plus E2E tests.

Use cases

  • Creating AI-generated comics and animated comics (manga) with consistent characters and structured storyboards.
  • Writing novels with AI-assisted outline generation, style control, and bulk chapter generation.
  • Building local knowledge bases with RAG, document/image vectorization, and knowledge graph visualization.
  • Running local AI dialogue with streaming output, deep thinking, and plugin skills.
  • Learning and experimenting with local AI inference pipelines on consumer hardware.

Getting started

See the repository README for installation and quick-start instructions.

License and pricing

The license is not stated in the repository metadata. The project is listed as open source and free to use.

Based on the project's public README and repository metadata.

Add this badge to your README

Featured on YunoDB Badge styles & help

Maintain this project? Show that it is featured on YunoDB.

Markdown[![Featured on YunoDB](https://yunodb.com/badge/yzw202011omnispace.svg)](https://yunodb.com/tools/yzw202011omnispace?utm_source=github&utm_medium=badge)
HTML<a href="https://yunodb.com/tools/yzw202011omnispace?utm_source=github&utm_medium=badge"><img src="https://yunodb.com/badge/yzw202011omnispace.svg" alt="Featured on YunoDB" height="20"></a>