NobodyWho
An efficient inference engine for running large language models locally offline
About NobodyWho
NobodyWho is an LLM inference engine for local deployment across multiple devices. It runs models locally offline without an API key or hidden fees. It provides type-safe tool calling and automatically generates structured syntax from function signatures, reducing the need to hand-write schema. It also supports multimodal inputs such as images and audio, as well as Kotlin, Swift, Python, Flutter, React Native, Expo, and Godot.
Best for
Best use cases
Run LLMs offline
Integrate local inference across platforms
Connect function tool calling
Process image and audio inputs
Features
Local offline execution
Run an LLM locally offline without an API key or hidden fees.
Type-safe tool calling
Automatically generate structured syntax from function signatures without hand-writing schema.
Multimodal input
Provide image and audio information to an LLM.
Multi-platform support
Supports Kotlin, Swift, Python, Flutter, React Native, Expo, and Godot.
Pros
- Run locally offline without an API key or hidden fees
- Supports Kotlin, Swift, Python, Flutter, React Native, Expo, and Godot
- Free to use in proprietary and commercial projects
Cons
- When distributing modified versions of the repository code, the corresponding modifications must be open-sourced
Pricing Plans
Free
- Usable in proprietary and commercial projects
* Pricing information is for reference only. Please refer to the latest announcement on NobodyWho's official website.