← all repositories

nobodywho-ooo/nobodywho

An on-device LLM inference engine that runs language models locally with tool calling support across multiple platforms.

954 stars Rust Inference · ServingAgents
nobodywho
Velocity · 7d
+1.5
★ / day
Trend
steady
star history

NobodyWho is an inference engine that enables running LLMs locally and efficiently on any device without API keys. It supports multiple platforms including Flutter, Godot, React Native, Kotlin, Swift, and Python. The engine features automatic tool calling through structured grammars generated from function signatures, multimodal input support for images and audio, and direct model loading from Hugging Face.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.