iiinova — where your digital self keeps evolving

NobodyWho @nobodywhooo.bsky.social · 14h

New NobodyWho release for Python, Flutter, Godot, Kotlin, Swift & React Native 🚀 - Godot bindings rewritten - OpenAI-compatible Python APIs - Configurable context shifting Check it out on docs.nobodywho.ai

000

NobodyWho @nobodywhooo.bsky.social · 05/10/2026

Running MiniCPM5 from OpenBMB on iPhone Air at 37 tok/s. One of the best open models at 2B parameters, now available in the NobodyWho Chat app.

MiniCPM5 from OpenBMB running on iPhone Air

100

NobodyWho @nobodywhooo.bsky.social · 01/10/2026

A 27B model in your pocket, running fully on-device, offline. Now available in the NobodyWho Chat app. iOS: apple.co/4cW30Jg Android: play.google.com/store/apps/d...

000

NobodyWho @nobodywhooo.bsky.social · 29/09/2026

NobodyWho in 25 seconds 🎬 Run open-weight AI models fully offline: 💬 Text generation 🖼️ Image + audio ingestion 🔊 STT, TTS & VAD 🔎 Embeddings & RAG Python, Flutter, RN/Expo, Swift, Kotlin, Godot... No cloud. No API keys. Open source. docs.nobodywho.ooo

021

NobodyWho @nobodywhooo.bsky.social · 23/09/2026

A new version of NobodyWho Chat has been released! 🎉 Includes a voice assistant, improved UI and new open-weight models. iOS: apple.co/4cW30Jg Android: play.google.com/store/apps/d...

ios and android app running AI models on-device offline

000

NobodyWho @nobodywhooo.bsky.social · 22/09/2026

Everyone and their mom is talking about Jev, here's the whole thing in 25 lines of Python www.nobodywho.ai/posts/jev-in...

000

NobodyWho @nobodywhooo.bsky.social · 17/09/2026

Running LLM inference on mobile means fighting the OS's memory manager. Our engineer Mads dug into applicationDidReceiveMemoryWarning on iOS and onTrimMemory on Android, and how to hook into them from a Rust library. 🔗 www.nobodywho.ai/posts/infere...

000

NobodyWho @nobodywhooo.bsky.social · 17/09/2026

NobodyWho just shipped a new release! → complete(messages): hand over a whole conversation → text, images and audio interleaved in one message → per-turn sampler & tools, ~400ms of overhead gone Python, Flutter, Kotlin, Swift, RN / Expo, Godot docs.nobodywho.ooo

docs.nobodywho.ooo

Home | NobodyWho

Local-first LLM inference for Kotlin, Swift, Python, Flutter, React Native, Expo and Godot

030

Reposted by NobodyWho

Snapp iOS Weekly @snappiosweekly.bsky.social · 04/09/2026

📢 𝗜𝘀𝘀𝘂𝗲 𝟭𝟬𝟴 𝗶𝘀 𝗼𝘂𝘁! Featuring: 🌐 𝑆𝑤𝑖𝑓𝑡 𝑊𝑎𝑠𝑚 📱 𝐵𝑢𝑖𝑙𝑑𝑖𝑛𝑔 𝐴𝐼 🎨 𝑇𝑒𝑎𝑐ℎ𝑖𝑛𝑔 𝑆𝑖𝑟𝑖 🚀 𝑀𝑎𝑘𝑖𝑛𝑔 𝑎𝑝𝑝 💡 𝐵𝑢𝑖𝑙𝑑𝑖𝑛𝑔 𝑠𝑢𝑛𝑏𝑢𝑟𝑠𝑡 🧠 𝐴𝑟𝑒𝑎 𝑐ℎ𝑎𝑟𝑡 🔧 093 ⚡ 𝑁𝑜𝑏𝑜𝑑𝑦𝑊ℎ𝑜 #Swift #iOSDev ios-newsletter.snappmobile.io/issues/2026-...

ios-newsletter.snappmobile.io

iOS Newsletter Issue 108

8 articles: Swift for Wasm August 2026 Updates, Building AI features using Foundation Models. Multimodal input., Teaching Siri to talk about your favorite walks in Walk Mate, and 5 more

142

NobodyWho @nobodywhooo.bsky.social · 31/08/2026

NobodyWho now works with Expo! 🎉 Run on-device LLMs in your Expo apps for free, no per-token bill. Build chat with tool calling or ship features like RAG, STT, TTS and VAD. Read the announcement and get the starter example : www.nobodywho.ai/posts/announ...

000

NobodyWho @nobodywhooo.bsky.social · 27/08/2026

Adding local, offline LLM support to your app? Let your coding agent handle it. The nobodywho skill gives your agent everything it needs 🙂 npx skills add github.com/nobodywho-oo... --skill nobodywho

000

NobodyWho @nobodywhooo.bsky.social · 25/08/2026

Just shipped a major release across all 6 bindings: Godot, Python, Flutter, Kotlin, Swift, React Native New features: - Voice Activity Detection - Batch embeddings - Faster tool calling Run LLMs + speech models fully on-device, no cloud. www.nobodywho.ai/posts/announ...

000

NobodyWho @nobodywhooo.bsky.social · 24/08/2026

How many CPU threads should you use for inference? Short answer: not "all of them." Turns out that's often the worst choice. We benchmarked across Ryzen, Apple Silicon, and mixed background load. Article: www.nobodywho.ai/posts/thread... Discussion on HN: news.ycombinator.com/item?id=4941...

000

NobodyWho @nobodywhooo.bsky.social · 20/08/2026

We are building a little offline city guide for iOS. It acts like a travel agent on your phone, showcasing what a small LLM can do combined with travel data from Wikivoyage.

100

NobodyWho @nobodywhooo.bsky.social · 20/08/2026

We just launched NobodyWho on Product Hunt 🚀 We’d really appreciate your support: check us out, leave a comment, and upvote if you like what we’re building! Link to support: www.producthunt.com/products/nob... Thanks ❤️

100

NobodyWho @nobodywhooo.bsky.social · 18/08/2026

Talk to your phone, get a voice back 😎 New voice assistant feature coming soon for NobodyWho Chat: - #Whisper for speech to text - LLM of your choise, here Granite - #Supertonic for text to speech Entire pipeline runs on-device. Offline

000

NobodyWho @nobodywhooo.bsky.social · 03/08/2026

NobodyWho Chat is live on iOS & Android! 🥳 A fully offline AI assistant. No internet, no accounts, no cloud. Run #Gemma, #Ministral, #Granite, #LFM, #Qwen and more, entirely on your device. Vision and hearing support included. Fully open-source. Links in comments 👇

222

NobodyWho @nobodywhooo.bsky.social · 14/07/2026

NobodyWho now supports Text-to-Speech & Speech-to-Text 🎉 Give your #LLM apps a voice and ears, fully on-device. No internet. No API keys. No usage fees. → Kokoro & Supertonic for TTS → Whisper for STT → 31+ languages → Open-source & free for commercial use github.com/nobodywho-oo...

010

NobodyWho @nobodywhooo.bsky.social · 16/06/2026

A new version of NobodyWho is out! 🎉   Support for Liquid AI's LFM2 & LFM2.5 (with tool-calling), getCachedModels() to list your downloaded models and their sizes, a seed parameter for reproducible sampling and some fixes.  📖 Docs: docs.nobodywho.ooo

docs.nobodywho.ooo

Home | NobodyWho

Local-first LLM inference for Kotlin, Swift, Python, Godot, Flutter, and React Native

020

NobodyWho @nobodywhooo.bsky.social · 11/06/2026

Picking your first local #AI model shouldn't be hard. We curated models from Google, Liquid AI, Mistral and other labs. Filter by what matters to you and find yours in seconds. Made a quick video showing how it works 👇 nobodywho.ai/models/

120

NobodyWho @nobodywhooo.bsky.social · 09/06/2026

Can't wait until fall for Siri AI on your Apple Watch? Or stuck in the EU? #WWDC Try NobodyWho Wrist: the world's first AI running fully on-device on Apple Watch. apps.apple.com/us/app/nobod...

000

NobodyWho @nobodywhooo.bsky.social · 04/06/2026

Releasing the world's first standalone AI assistant on Apple Watch working fully offline. Computing power is limited on Apple Watch, so we focused on small but efficient models from PrismML and LiquidAI. We’ve also published a Vision Pro app with larger models. www.nobodywho.ai/posts/apple-...

000

NobodyWho @nobodywhooo.bsky.social · 03/06/2026

NobodyWho now supports #Kotlin 🎉 Run #LLMs fully on-device in your native #Android apps. No internet. No API keys. No usage fees. → #Gemma4, Qwen3.5 & more (.gguf) → Hardware acceleration → Tool calling, RAG, vision & audio ingestion github.com/nobodywho-oo...

000

NobodyWho @nobodywhooo.bsky.social · 02/06/2026

“#LLM, give me a #JSON. Make no mistakes." Spoiler: it makes mistakes. We dive into why prompt-based JSON is unreliable, how constrained sampling works, and how we integrated LLGuidance so you can run structured, reliable inference locally on any device. www.nobodywho.ai/posts/llm-gi...

000

NobodyWho @nobodywhooo.bsky.social · 19/05/2026

NobodyWho now supports #Swift 🎉 Run #LLMs fully on-device in your #iOS, #macOS, #watchOS & #visionOS apps. No internet. No API keys. No usage fees. → #Gemma4, #Qwen & more (.gguf) → Hardware acceleration → Tool calling, #RAG, vision & audio ingestion → Open-source & free github.com/nobodywho-oo...

000

NobodyWho @nobodywhooo.bsky.social · 15/05/2026

Open standards are important to us at NobodyWho, and GGUF is a great one. Of course GGUF files contains the weights for an #LLM, but there's also a bunch of other stuff. What's all the other stuff, and is it sufficient? I wrote a blogpost about it. www.nobodywho.ooo/posts/whats-...

010

NobodyWho @nobodywhooo.bsky.social · 12/05/2026

NobodyWho just got a major structured output upgrade! 🎉 Constrain #LLM responses with JSON Schema, regex or full context-free grammars, and it's faster too. Available for #Flutter #ReactNative #Godot and #Python Docs: docs.nobodywho.ooo

240

NobodyWho @nobodywhooo.bsky.social · 07/05/2026

#AI has its own language. If you're just getting started, it can feel like everyone else got the memo but you 🤯 This guide helps you navigate the AI jungle ⬇️ www.nobodywho.ai/posts/beginn...

Essential Terms in Artificial Intelligence

010

NobodyWho @nobodywhooo.bsky.social · 29/04/2026

A new version of NobodyWho is out! 🎉 Remote model downloads now report progress — plus a GPU memory bug fix improving inference speed by up to 30× 📖 Docs: docs.nobodywho.ooo

010

NobodyWho @nobodywhooo.bsky.social · 27/04/2026

Today we're announcing our #ReactNative bindings for NobodyWho! 🎉 Run #LLM models offline in your #react apps for #Android and #iOS. No internet. No API keys. No usage fees. Fully open-source. www.nobodywho.ai/posts/announ...

051

NobodyWho @nobodywhooo.bsky.social · 20/04/2026

We are releasing multimodal support for #Flutter, #Python, #Godot and model downloading from @hf.co ! Also dropping: support for Qwen 3.5 & 3.6, Ministral 3 and Gemma 4, and automatic sampler settings from GGUF metadata! Flutter: pub.dev/packages/nobodywho Python & Godot: docs.nobodywho.ooo

010

NobodyWho @nobodywhooo.bsky.social · 09/04/2026

Running a 1.7B parameter #LLM on an Apple Watch thanks to PrismML 1-bit Bonsai ultra-dense intelligence. Fast, fully on-device, works offline. github.com/nobodywho-oo...

030

NobodyWho @nobodywhooo.bsky.social · 30/03/2026

Apple Watch still can't run Apple Intelligence. So we did it ourselves 👀 Run open-weight LLMs from huggingface. Fully offline. On your wrist. Powered by the open-source NobodyWho engine: github.com/nobodywho-oo... More details dropping soon!

010

NobodyWho @nobodywhooo.bsky.social · 30/03/2026

What if your #Flutter app had #AI features that worked offline, cost nothing to run, and never sent user data to the cloud? That's #OnDeviceAI and it's easier than you think. New guide: #LLM in Flutter with NobodyWho, from basic chat to RAG 👇 dev.to/pielounw/run...

030