Suggestions appear as you type. Use the up and down arrows to choose one and Enter to open it.

OnDevice LLM review

Free#34 of 66 in Local LLM Software

A free, open-source local AI assistant with APIs, agents, and device pairing.

5.8/10Editor score
OnDevice LLM5.8 Visit OnDevice LLM

Reviewed by iTechGuides Editors · Editorial team · Updated Nov 2026

OnDevice LLM is a local-first assistant for Android phones and Windows PCs. It runs GGUF language models on-device through llama.cpp and supports chat, document and photo analysis, voice conversation, local agents, model discovery, and downloads. The project suits solo users and small teams that want local AI without an account or subscription. Local data stays on the device unless networking features are enabled, while the API server can expose models to compatible applications.

As an open-source project with a free plan, OnDevice LLM keeps access straightforward rather than dividing capabilities across paid tiers. Its ecosystem fit is practical: the local API supports OpenAI- and Anthropic-compatible clients, and Hugging Face discovery extends the built-in model catalog. Android and Windows are the primary user platforms, with API access for integrations. Hardware options include CUDA, Vulkan, OpenCL, and CPU acceleration, giving users several ways to match inference to their device.

The feature set reaches beyond basic chat. Local agents can run parallel tasks, while photo, PDF, and document analysis, on-device speech, text-to-speech, cited web research, and multi-step research Goals cover varied workflows. Phone-to-PC pairing enables remote local inference, and the Windows build adds image and video generation, model comparison, and optimization settings. The trade-off is scope: pairing is centered on Android and Windows, and those generation features are Windows-specific. Users seeking broader platform coverage or vendor-managed support should consider an alternative; users prioritizing local control and an open-source workflow should find this a focused fit.

OnDevice LLM pros and cons

  • Where it wins
    • Runs GGUF models locally with CUDA, Vulkan, OpenCL, or CPU acceleration
    • OpenAI- and Anthropic-compatible local API server
    • Combines agents, document analysis, voice, research, and model downloads
  • Where it doesn't
    • Phone-to-PC pairing centers on Android and Windows devices
    • Image and video generation features are available on Windows
    • Support is provided through community channels and documentation

OnDevice LLM fact sheet, pricing and score →

Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.

Last updated · How we research and update