I found these scratch post it notes at the store and thought it would be good material for the plotter. Super fun! . #generativeartworks #generativeart #generative #creativecoding #codeart #artandtechnology #artxcode #computerart #creativecodeart #axidraw #plottertwitter

LM Studio featured in Apple's new M5 announcement! 🥳🚀

Apple just won the AI war They announced the MacBook Pro M5 Max What would have cost $40,000 to do with Nvidia chips, now costs $3,000 with a laptop • 4x faster AI speeds • Up to 128gb of unified memory • Frontier intelligence on the go The biggest complaint about Apple

Xcode 26.3 with Claude Agent & Codex hits the Mac App Store today! With advanced reasoning capabilities in Xcode, you can streamline workflows and build faster. And MCP support lets you easily connect other compatible agents.

Pretty incredible that this is running 100% locally on 2 x 512GB M3 Ultra Mac Studios connected with a Thunderbolt 5 cable consuming ~400W. This is possible now because of a perfect storm of: - Models: Really good Chinese open models. - Hardware: Apple Silicon with unified

Do you even understand what this means? An open source model just released that is: • Just as smart as Sonnet 4.5 • Incredible at coding • Can run on almost any modern computer If you have 32gb of RAM (most Mac Minis do) you can have unlimited super intelligence on your

https://t.co/JJYRU4BG0b

Claude throwing API errors AND billing me for extra usage when I'm within my session limit on Max. @claudeai you good?

Anne Spalter @annespalter

Presenting my Mac Mini Cluster of 3!!! So here is a pic of my small Mac mini cluster of 3 M4 16 gigs RAM 2025 editions rotating 48 gigs of unified memory All of them connected thunderbolt 4 which gives me around 10 gbps All of them them has the standard 1 gbps Ethernet port

Your Mac can host a cluster of macOS VMs, each running a Clawdbot server. Thanks @steipete for merging Lume support into the docs — run @openclaw in sandboxed macOS VMs on hardware you already own. iCloud included. https://t.co/bKVHRz0m3y

more than 12000 visitors saw 486 Shorts at OFFSCREEN Paris last week. We had a featured magazine review, photos in numerous publications and the most IG stories traffic we've ever had. One edition sold, conversations with curators from from around the world

Great start to the year of local AI. GLM-4.7-Flash feels competitive with frontier coding models from ~a year ago and runs at >80 tok/sec on a MacBook.

this is huge. ollama is now compatible with the anthropic messages API. which means you can use claude code with open-source models. think about that for a second. the entire claude harness: - the agentic loops - the tool use - the coding workflows all powered by private LLMs

The latest mlx-lm is out with some nice new features: - Support for GLM 4.7 flash (@ivanfioravanti) and iQuest coder loop (@kernelpool) - Support autoAWQ and autoGPTQ models (@ericcurtin17) - General improvements to continuous batching and distributed inference

this is my life now

I have a personal rule that the price of being distracted by a new side-project is that I have to write about it Claude Code over the past ~4 weeks has caused me to break my rule at least half a dozen times, I am SO behind on writing up my projects right now!

Vibe code project of the week: A soundboard of my toddler's cutest words $20 mic off Amazon + Descript to slice up each word + Claude Code to clean up the audio quality + Lovable to build

Who knows what 2026 will bring! Happy New Year from outer space, the dessert, and the art world 🎉🛸🌪🌵

It genuinely feels to me like GPT-5.2 and Opus 4.5 in November represent an inflection point - one of those moments where the models get incrementally better in a way that tips across an invisible capability line where suddenly a whole bunch of much harder coding problems open up

First release of mlx-lm in 2026 is packed: pip install -U mlx-lm - Bunch of new models (h/t @kernelpool, @JohnMai_Dev) - Much better support for tool calling and reasoning in mlx_lm.server - Support for mxfp8 and nvfp4 quantization (require pre-release mlx)