AiOS Dispatch 7
Welcome to the seventh edition of AiOS Dispatch, your guide to the AI advancements, iOS development, and the tools shaping our workflows.
I am typing this from a cozy cafe in Yokohama, chilling after try! Swift Tokyo, where I got to talk to many Japanese and Taiwanese developers! It was interesting to know how they use AI assisted tools in their workflow and various "AI" policy in bigger organizations.
It is a Sunday, so I will keep this one short, talking about two models: the mysterious Quasar Alpha and Optimus Alpha models on OpenRouter!
Stealth Models
OpenRouter is my go-to service for using LLM APIs as it provides simple OpenAI SDK based endpoints and model ID, making it much easier to test different models.
If you hang around this website you might have noticed two new "stealth" models appearing this month: Quasar Alpha and Optimus Alpha. While I cannot see a way to use Quasar Alpha anymore, both the models come with 1M context and free to use. Caveat? All prompts and completions will be logged for the lab to understand how it is being used and where it can improve.
There are strong hints and technical breadcrumbs (like API response formats) suggest they might be unannounced previews from OpenAI, perhaps related to the GPT-4.1/4.5 family or the rumoured o4-mini
or even the open-source model that they have been talking about. Sam Altman even tweeted cryptically about "quasars" around the time the model appeared!
Also, a speculative tweet about its implementation:
https://x.com/shaunralston/status/1910711627772813758
Optimus Alpha
I have been testing this 32K max output model via Alex Sidebar and I like it. I do not think this will be my default model because Gemini 2.5 Pro is still better for me, but it is a decent contender for a general-purpose model.
For iOS work, the faster code generation is fantastic for quick prototyping or utility functions.
The amount of tokens/day usage in tens of billions signals massive developer interest. But again, it is available for free, and the cloaked lab is collecting data to improve the model. If that is something that does not bother you, give it a try!
Moving Forward
Almost done with my vacation and I miss the grind of working 12+ hours a day. I am working on a new book on MLX Swift + App Intents (later on but before WWDC 2025) and hyped about getting it in the hands of developers.
Expect this dispatch twice a week to cover in more depth and more news that are tailored for you as an iOS/macOS developer.
Happy exploring!