KernelAI: 53 open models from 15 labs, running entirely on your phone
KernelAI is a free iPhone/Android app launching today that runs 53 open-weight models (112 MB to 3.11 GB) from 15 labs fully on-device — chat, PDF/spreadsheet RAG, vision, offline coding, on-device dictation, Shortcuts actions, and experimental GGUF import.
Why it matters
A polished, no-account, no-subscription mobile client that turns the phone into a private multi-model lab — on-device RAG and GGUF import make it a real prototyping tool, not just a chat wrapper.
What you could build with it
A mobile startup prototyping offline-first AI features could use KernelAI's 53-model catalog to benchmark on-device quality for document chat, vision, and code tasks before committing to a custom llama.cpp integration.
Does it hold up?
Too early to judge — launched on Product Hunt today; the feature list (RAG, vision, code, no account, optional Tavily/Exa web search) is all from the maker so far, with no third-party hands-on reviews yet.
Learn more
First spotted on producthunt: source.
More AI models
OpenAI publishes 722 AI-generated math manuscripts from an unreleased modelOpenAI released 722 mathematical manuscripts in 372 result families to GitHub under Apache-2.0, produced by an…models · JEV 0.74Mistral Large 4 (Le Chonk): 1.05T-param open-weight MoEMistral's new flagship: a 1.05T-parameter mixture-of-experts multimodal model with 49B active params and 1M-token…models · JEV 0.72ElevenLabs Eleven v4 + v4 Turbo: new emotive TTS architecture with a 100ms real-time variantElevenLabs released Eleven v4 and Eleven v4 Turbo on September 28: a new text-to-speech architecture with inline…models · JEV 0.72VoiceStudio: open-source, fully-local ElevenLabs alternative with voice cloning, dubbing and dictation in 646 languagesVoiceStudio is an open-source, fully-local voice AI studio: voice cloning from a 3-second sample, voice design, video…models · JEV 0.71JetBrains Mellum 2.1: 12B MoE coding model trained for agentsJetBrains released Mellum 2.1, a 12B MoE model (2.5B active parameters, 128K context) under Apache 2.0, trained mainly…models · JEV 0.69DeepSeek V4-Flash official API: public beta with upgraded agent capabilities, Responses API and Codex supportDeepSeek's official V4-Flash is now a public beta API with massively upgraded agent capabilities - benchmark scores…models · JEV 0.69
Get the week's best AI launches, plus 3 ideas worth building
One email every Saturday. Ranked by traction, not hype. Free.