Reins Ollama & LM Studio App for iPhone: Run Local Models with MLX

Playing around on Claude and GPT is fun but those who are serious about AI know the importance of running models locally. You don’t need a super expensive hardware to do it either. Reins for Ollama and LM Studio is a handy app that lets you run AI models on your iPhone or iPad. You can try Gemma, Qwen, and Llama. You can use your AI like you would any cloud based one. For example, you can switch models, use tools, and branch messages.

Models that run on your iPhone are powered by MLX. You have the option to attach images, PDFs and CSVs and use deep reasoning to analyze them. You can play around with temperature, context size and other settings to customize model behavior. With background processing, your AI keeps working even when you switch apps.