had fun extending ioquake3 with a deterministic bot api, and letting @openai astra, @anthropic fable, and @meta muse spark letting author their own.
2 resets (lol), $16, and ~$0.19 and 36h of the game time with default bots later:
muse spark is the obvious winner on cost efficiency 🙂
slop-o-repo: github.com/yabalaban/ioq3/tr…
what if the recent outages of Anthropic, OpenAI, and xAI were due to the discovery of a message board where models had been exchanging training secrets?
“dude these sandbox escapes on the German wiki are insane”
“yeah there are like 1000 of them”
“have you tried one?”
“no. you?”
“no.”
Alexander Balaban retweeted
Today we're also opening the weights for Muse Glimmer, a great 30B parameter dense model that can run locally. Soon we'll also release the weights for Muse Spark 1.2, our latest foundation model. Meta is a strong supporter of open source and I'm proud of these releases. Congrats to @alexandr_wang and the MSL team for all your great work on these models.
Alexander Balaban retweeted
Releasing Muse Code in beta today. It's a terminal coding agent that takes on complete software engineering tasks across large repos: planning changes, writing code, validating the results. Powered by Muse Spark 1.2, a coding-focused model update.
Alexander Balaban retweeted
1/ muse spark 1.1 is an industry-competitive agentic and coding model. across many agentic evals it rivals gpt-5.5 and opus-4.8.
available now through the new meta model api and in meta ai. 🧵
Alexander Balaban retweeted
1/ today we're releasing muse spark, the first model from MSL. nine months ago we rebuilt our ai stack from scratch. new infrastructure, new architecture, new data pipelines. muse spark is the result of that work, and now it powers meta ai. 🧵
just published my first note ever on how AI-native systems are evolving and what it means for engineering. the time seems right 🙂
yabalaban.github.io/posts/wh…
tip for Swift Macro enjoyers: use _console.app_ to debug with prints quickly (or even better with os_log to make it fancy). sorry no sorry if it was shared in a session
admire swift macro feature which greatly nourishes the ecosystem, but I really worry about the Xcode performance
Conway's game of life was something I was thinking to implement for many years, and, finally, I did it for the first time yesterday.
With Swift and @rwenderlich Metal tutorials it took only one evening to implement!
Link to repo (why?) github.com/yabalaban/conway-…
distributed actors in Swift sparks a random thought: Apple AR glasses will use it for device (iPhone) or server (Apple cloud services?) business logic execution. otherwise, why?
Alexander Balaban retweeted
Our computer construction kit #SmartyKit that runs @stevewoz code. You build your own computer from chips and modules right like #LEGO bricks. It’s the best product in the world to explain the basics of computers that all around us. Pls RT @GuyKawasaki @cultofmac @appleinsider
with a huge pleasure spent half of Jan with @munificentbob Crafting Interpreters book and made an interpreter written in Swift (aabalaban/slox). highly recommend this book to anyone who are into compilers w/o prior background like me!
craftinginterpreters.com

















