LLM Inference on Edge: A Fun and Easy Guide to run LLMs via React Native on your Phone!
Hugging Face Blog
Developers can now run large language models directly on smartphones using React Native by downloading models from Hugging Face and executing them locally with llama.rn bindings. The tutorial uses models like DeepSeek R1 Distill Qwen with 1.5 billion parameters, which fit within mobile device constraints when properly quantized. Users gain a fully offline, privacy-preserving chat application compatible with both Android and iOS that requires no cloud connectivity.
