Otaku — an LLM frontend
- 发布
- 09/05 16:29
- 收录
- 09/05 20:00
- 来源类型
- 开发者社区
- 档位
- 社区
- 信源状态
- 正常
Otaku 是一款专为角色扮演设计的 LLM 前端，可作为 SillyTavern 的替代品。它也支持与本地后端（如 Ollama、oMLX、LM Studio、llama.cpp 和 KoboldCpp）或云模型（如 OpenRouter、NanoGPT）进行通用聊天。首次启动时，用户可以选择提供商和模型，Otaku 会自动检测本地安装。用户可以使用 /new 命令开始新的故事，开发者正在积极征求社区反馈以进行产品改进和功能添加。
Otaku is an LLM frontend, primarily designed for roleplay, an alternative to SillyTavern and the like. However, It also works for general-purpose chat with local backends (including Ollama) or cloud models, the way Open WebUI is used, once lore extraction is switched off in the settings.
Both share the same functions; the difference is that in the terminal you execute them with slash commands (the reference is available with /help), while in the web UI the operations are available from the menu.
Otaku is free and open source (MIT); it works on macOS, Linux and Windows. Install it with uv (uv tool install otaku) or see the GitHub README for other options: https://github.com/enclavum/otaku
Launch either otaku for the terminal or otaku web for the web UI; the web UI's default URL is http://localhost:9600. Two sample stories are imported on first start to give you an idea of the features and what play looks like, and you land right in the middle of one of them.
On first start, you choose a provider and a model: Otaku automatically detects local installations of Ollama, oMLX, LM Studio, llama.cpp and KoboldCpp, and lets you pick from their models. Cloud providers (OpenRouter, NanoGPT) are also there: enter an API key and their catalogs appear. After exploring the provided stories, you can start your own with the /new command.
Otaku is a personal side project, and I'd like to get feedback from the community on the product and on what to add.