Categories
AI Agents Self-Hosted AI AI Chat Local LLM LLM Frontend AI Game Master AI Roleplay Character CardsStars
Forks
Watchers
Developer links
Marinara Engine
Marinara Engine turns any AI model into a character chat, a roleplay stage, or a tabletop-style game run by an AI Game Master, all in one self-hosted web app. Conversation mode feels like a messenger, with group chats, audio and video calls, selfies, and characters who keep daily schedules and message you first. Roleplay mode adds a stage behind the words: character sprites, scene backgrounds, weather effects, and trackers for time, quests, stats, and mood. Game mode hands the story to a Game Master that runs NPCs, dice and skill checks, classic or tactical battles, maps, and saved sessions on rulesets such as 5e (SRD 5.1). Connect OpenAI, Anthropic, Gemini, OpenRouter, DeepSeek, or any OpenAI-compatible endpoint, run local models through Ollama, LM Studio, or KoboldCpp, or sign in with a Claude, ChatGPT, or Grok plan instead of an API key. Optional agents work beside each chat to check continuity, track the world, draw illustrations, play music, and recall long-term memory, while Professor Mari, the built-in assistant, helps with setup. Lorebooks add world facts when keywords appear, and Peek Prompt shows exactly what the model received. Import characters, chats, and lorebooks from SillyTavern, then export chats as JSONL, Markdown, or styled HTML. Visitors from the network or internet must sign in, with Basic Auth and IP allowlists for extra control. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. AGPL-3.0 licensed.
Benefits
- Three Play Modes In One App
- Switch between messenger-style Conversation, sprite-driven Roleplay, and Game mode with an AI Game Master, and let linked chats share context so one cast of characters serves every style of play.
- Bring Any Model Or Plan
- Use cloud APIs, local Ollama, LM Studio, or KoboldCpp servers, or a Claude, ChatGPT, or Grok subscription login, and pick a different model for each chat without reconfiguring anything.
- Keep All Your Data Local
- Characters, chats, lorebooks, and settings live in a data folder on your own server, API keys are encrypted with AES-256, and automatic backups can run daily, weekly, or monthly.
- Control Every Prompt Detail
- Drag-and-drop presets, keyword lorebooks, macros, regex scripts, and Peek Prompt let you see and change exactly what the model receives, which keeps long stories consistent and easy to debug.
Features
- Three Chat Modes
- Conversation, Roleplay, and Game modes share characters, memory, branches, and group chats, with sprites, backgrounds, trackers, dice, maps, and tactical combat available where each mode needs them.
- Wide Model Connections
- Connects to OpenAI, Anthropic, Google Gemini, Vertex AI, OpenRouter, Mistral, xAI, DeepSeek, and any OpenAI-compatible endpoint, plus Ollama, LM Studio, KoboldCpp, llama.cpp, and a built-in local helper model.
- Downloadable AI Agents
- An Agent Organizer installs optional helpers such as Prose Guardian, Continuity Checker, World State, Illustrator, and Music DJ from the official Marinara-Agents catalog, and you can build custom agents.
- Professor Mari Assistant
- A built-in assistant on the Home screen helps configure connections, draft lorebooks and themes, and make changes that stay reversible, so first-time setup needs no manual file editing.
- Media And Voice Services
- Image generation through OpenAI, NovelAI, Stability AI, fal.ai, or local ComfyUI, text to speech through ElevenLabs or OpenAI-compatible servers, and Music DJ playback from Spotify, YouTube, or local files.
- Docker And PWA Deployment
- The official ghcr.io image runs with docker compose on port 7860 with a persistent data volume, and the responsive interface installs as a Progressive Web App on phones and tablets.