Private LLM

Local AI for Private, Uncensored Chat on iPhone, iPad, and Mac. No Cloud, No Tracking, No Logins.

[Private LLM @private_llm]

Can an iPhone hold its own against a Mac? 💻📱

Ollama is running DeepSeek R1 Distill Llama 8B on a 16GB M2 MacBook Air, while the iPhone 15 Pro (8GB A17 Pro) runs the same model.

Same model. Same sampling temperature. Same system prompt.

While the iPhone is a bit slower, the 3-bit OmniQuant model reasons just as well as the 4-bit RTN model on the Mac. 🤯

[Private LLM @private_llm]

There is no such thing as “private” chats with ChatGPT

[Rob Freund @RobertFreundLaw]

ChatGPT allegedly shares your chat query topics, user IDs, and email addresses with Google and Meta, according to a new class action lawsuit filed today.

[Private LLM @private_llm]

Only a concern if you’re using unsandboxed apps like Ollama and LM Studio for local LLM inference. IYKYK. 💀

[The Hacker News @TheHackersNews]

⚠️ Attackers poisoned Hugging Face & ClawHub (OpenClaw) with 575+ malicious skills from just 13 accounts.

🔸 Fake helpful AI tools that install trojans, miners & stealers (Windows + macOS)
🔸 Use hidden commands & indirect prompt injection

Quick action: Never install random AI skills or models. Always verify the source.

Read: https://t.co/CmdDBXuzTy

[Private LLM @private_llm]

@philliplanos All QWen 3.5 and Gemma 4 models support speculative decoding with MTP (Multi Token Prediction). We’re working on adding support in PLLM2, RN. 🔜

[Private LLM @private_llm]

@pmarreck 🔜

[Private LLM @private_llm]

@shanutweets QWen 3.6 models are not out yet.

As announced on our discord, we’re working on Private LLM 2.0 which is a new app with many unprecedented features (free upgrade for Private LLM 1 users). Will out in ~3 weeks. Gemma 4 and QWen 3.5 models will be in it.

[Private LLM @private_llm]

@rgj972690 Coming in Private LLM2. See our discord.

[Private LLM @private_llm]

@BadTechBandit As announced on our Discord, coming in Private LLM 2.

[Clean Links @CleanLinksApp]

Coming soon in the next update to Clean Links for macOS. Spot and clean tracking URLs on any website. Works fully offline. Also available on iOS.

[Private LLM @private_llm]

@olii_dev Why lie?

[Private LLM @private_llm]

Been shipping one on the App Store for nearly 3 years now. 😭

[Seth Fowler @sethlfowler]

@jack Local LLM on iPhone will be wonderful

[Private LLM @private_llm]

@Michael_IS_FI Did you miss last month’s update?

[Private LLM @private_llm]

@olii_dev We don’t handle refunds, Apple does. Did you miss last month’s update?

[John Scott-Railton @jsrailton]

Are you talking to AI about legal matters? Be careful.

He chatted with Claude about his legal strategy.

Then discussed with his lawyers.

A Federal judge just ruled that those AI chat logs have zero legal privilege

[SlashGear @slashgear]

AI chatbots are increasingly popular, but several users have privacy concerns. If you want to avoid the risk, there are ways to run your own locally. https://t.co/rmW1Gt3LMi

[SlopOrNotAI @SlopOrNotAI]

Our AI content detection model is small enough to run on Apple Neural Engine and works without watermarks or signature

[Poonam Soni @CodeByPoonam]

The Nano Banana 2 leaks are insane.

You can’t even tell what’s real anymore.

But I found this AI content detector that instantly flags Nano Banana 2 images.

(Meet Slop or Not👇)

[Private LLM @private_llm]

Link to the paper:
https://t.co/H3WMRrmZka

[Private LLM @private_llm]

Just learnt that Private LLM has been cited in an AI safety paper. Because we let users download and use lots of uncensored models.

[Private LLM @private_llm]

https://t.co/0ow4q813Df

[Private LLM @private_llm]

PSA: Non-native electron.js based apps cause systemwide lag in macOS 26 Tahoe. And many popular local LLM apps use electron.js. Private LLM isn’t one of them, because it’s fully native (and sandboxed) app. Link to the script in the tweet below. 👇