FeedCity

It's been an extremely busy day for team Qwen. Within the last 24 hours (all links to Twitter, which seems to be their preferred platform for these announcements): Qwen3-Next-80B-A3B-Instruct-FP8 and Qwen3-Next-80B-A3B-Thinking-FP8 - official FP8 quantized versions of thei...

CompileBench: Can AI Compile 22-year-old Code? Interesting new LLM benchmark from Piotr Grabowski and Piotr Migdał: how well can different models handle compilation challenges such as cross-compiling gucr for ARM64 architecture? This is one of my favorite applications of cod...

ChatGPT Is Blowing Up Marriages as Spouses Use AI to Attack Their Partners

Maggie Harrison Dupré for Futurism. It turns out having an always-available "marriage therapist" with a sycophantic instinct to always take your side is catastrophic for relationships.

The tension in the vehicle is palpable. The marriage has been on the rocks for months, and the wife in the passenger seat, who recently requested an official separation, has been asking her spouse not to fight with her in front of their kids. But as the family speeds down the roadway, the spouse in the driver’s seat pulls out a smartphone and starts quizzing ChatGPT’s Voice Mode about their relationship problems, feeding the chatbot leading prompts that result in the AI browbeating her wife in front of their preschool-aged children.

Tags: ai, generative-ai, chatgpt, llms, ai-ethics, ai-personality

Today's links It's still censorship (even if it doesn't violate the First Amendment): Communications monopolies create single points of failure. Hey look at this: Delights to delectate. Object permanence: Good emails; Dieselgate DRM; Precursor; London cops mug blogger; RFID c...

I was reading Chase McCoy’s article “Antibuildings” where he cites Wikipedia’s entry on the term “Antilibrary” which points to another entry about the Japanese concept of Tsundoku, all of which deal with this idea of things we do with intention but that never make it to frui...

Locally AI

Handy new iOS app by Adrien Grondin for running local LLMs on your phone. It just added support for the new iOS 26 Apple Foundation model, so you can install this app and instantly start a conversation with that model without any additional download.

The app can also run a variety of other models using MLX, including embers of the Gemma, Llama 3.2, and and Qwen families.

Tags: apple, ios, ai, generative-ai, local-llms, llms, mlx

TLDR; You can find the project repository here. What? How? Why? A few Christmases ago, I was browsing the source code for the esp-idf heap allocator1 and thought: This is quite interesting, I should write my own allocator After a bit of looking around, I discovered that general purpose heap allocators are one of those problems that has no perfect solution (which is the kind of problem I really enjoy). It’s all about trade-offs.

Imagine jumping on board a class action after your precious datas have been breached, then sticking through it all the way until a settlement is reached. Then, finally, after a long and arduous battle, cashing in and getting... $1. Well, kinda $1, the ParkMobile class action granted up to $1

Net uitgelezen, het handzame boekje “Het kon minder, de lessen van Johan Remkes” (bol.com e-book). Geschreven door Thijs Broer en Peter Kee die samen toch al heel wat van de politieke wereld gezien en besproken hebben. Het boek is geen biografie, verwacht geen stapels voetnoten en analyses. Maar dat belooft de intro ook al niet. “‘Een boek over mij, dat hoeft niet zo nodig,’ zei Remkes. Pas toen we hem verzekerden dat het geen biografie moest worden, maar dat we wilden beschrijven hoe hij het had gebracht tot veelgevraagd probleemoplosser in dit stuurloze land […] en wat anderen daar misschien van konden leren, pas toen bromde hij: ‘Laten we dan maar eens praten’”.

I liked the NixOS meetup earlier this year, and at the end of the meetup they told everyone about NixCon 2025, which would be happening in Switzerland this year, at the very same location, the University Of Applied Sciences OST in Rapperswil, so I decided to go! In this trip...

llm-openrouter 0.5 New release of my LLM plugin for accessing models made available via OpenRouter. The release notes in full: Support for tool calling. Thanks, James Sanford. #43 Support for reasoning options, for example llm -m openrouter/openai/gpt-5 'prove dogs exist'...

Grok 4 Fast New hosted reasoning model from xAI that's designed to be fast and extremely competitive on price. It has a 2 million token context window and "was trained end-to-end with tool-use reinforcement learning". It's priced at $0.20/million input tokens and $0.50/milli...

Amazonians, We've reviewed the Presidential Proclamation on H-1B visas that was released today and are actively working to gain greater clarity. Here's what you need to know right now: The proclamation creates a travel restriction starting September 21, 2025, at 12:01 a.m. ...

tl;dr: A meandering talk that reminds us of the risks of shipping unknown build time dependencies (as from npm), and also highlights how many services now have runtime third party service dependencies. These are very serious choices about the future of a project, and we can wonder how well we are making these choices. Is anyone thinking about the long term costs and impact? The talk then covers “microstacks”, whereby a project relies in a minimal number of build dependencies, and no runtime service dependencies.

httpjail Here's a promising new (experimental) project in the sandboxing space from Ammar Bandukwala at Coder. httpjail provides a Rust CLI tool for running an individual process against a custom configured HTTP proxy. The initial goal is to help run coding agents like Claud...

The Hidden Risk in Notion 3.0 AI Agents: Web Search Tool Abuse for Data Exfiltration Abi Raghuram reports that Notion 3.0, released yesterday, introduces new prompt injection data exfiltration vulnerabilities thanks to enabling lethal trifecta attacks. Abi's attack involves ...

Mistral quietly released two new models yesterday: Magistral Small 1.2 (Apache 2.0, 96.1 GB on Hugging Face) and Magistral Medium 1.2 (not open weights same as Mistral's other "medium" models.) Despite being described as "minor updates" to the Magistral 1.1 models these hav...

You can finally manage Macs with FileVault remotely in Tahoe Jeff Geerling September 19, 2025

I regret building this $3000 Pi AI cluster Jeff Geerling September 19, 2025