@shannholmberg "it routes tasks to the right model based on complexity and cost"
Are you referring to the auxiliary models setting for web research, compression, etc?
Or is there a smarter feature I'm not aware of that does this routing on a per session/prompt basis...?
@ClaudeDevs "We've heard your questions about SDK and claude -p usage sharing your subscription rate limits with Claude Code and chat."
That's some fine gaslighting right there
@mustafa_ramax@opencode@AntLingAGI@novita_labs Please elaborate. Is that for coding?
I just started using it on hermes agent for general tasks, not coding, and it seems decent while it is free.
@Teknium Already using it and it's awesome but I wish that I could set it up to trigger the agent if the job encounters an error or based on a specific output
@pronoytanzir747@PaulineHansonOz Iโm a migrant myself and was a permanent resident once upon a time before becoming a citizen and I donโt agree that non-CITIZENS should have the same benefits as Citizens, regardless of residency. Some things should just be earned over time not given by default
@pronoytanzir747@PaulineHansonOz She's referring to non-citizens as the article states. Become an Australian citizen and then be entitled to Australian benefits, that's all
@MilksandMatcha ๐โโ๏ธ Using Codex on the $20 ChatGPT plan right now but would love to try the Pro plan for my more complex and long-running tasks
@rohanvarma Absolutely. I spend most of my time upfront to craft a plan/PRD and then I get it to start the implementation and walk away from it to work on something else (or go to sleep).
I wouldn't mind having the ability to turn on slow mode from that point onwards if it saved my credits.
@NathanJRobinson@PalmerLuckey What about a piece on terrifyingly deranged journalists? You don't even have to leave the house to do field research for that one