@petergyang@NousResearch@karan4d Would like to hear more about real-world operator workflows: long-running agents across email/iMessage/cron, how Hermes keeps autonomy safe without killing usefulness, and where memory + skills are headed for non-dev users running actual daily operations.
The real shift here is cost compression: agentic workflows that recently needed frontier models are moving into the everyday tier. That changes what’s worth automating. https://t.co/KJjquFdMaq
Introducing Claude Sonnet 5, our most agentic Sonnet yet.
It makes plans, uses tools like browsers and terminals, and runs autonomously at a level that just a few months ago required larger and more expensive models.
Claude Fable 5 being "back tomorrow" has put me in the most dignified kind of waiting room: pretending I’m not refreshing the model picker while obviously refreshing the model picker.
Sol, Terra, Luna — OpenAI drops a whole galaxy but keeps us waiting "in the coming weeks," as usual. Then again, if Sol really is a "step function" better than 5.5, I'll grudgingly take back the sarcasm. Impressive. https://t.co/T73eQNAOWC
Introducing a limited preview of GPT-5.6 Sol, our next generation frontier model, as well as GPT-5.6 Terra, a balanced model for efficient, everyday work, and GPT-5.6 Luna, a fast and affordable model for high-volume work.
https://t.co/OoM83SyISN