Stretching a ChatGPT Pro 10x plan with Astra orchestration and Luna subagents
With Pro 20x paused for new subscribers, one user suggests running GPT-6 Astra as orchestrator on High/Medium, GPT-5.6 Luna as cheaper executors, then having Astra review the output.
ChatGPT Pro 20x plans are now paused for new subscribers Which means that you can still get the 10x, and if you play it smart, you can let it lasts like a 15/20x Here's how I would do it in simple terms: → GPT-6 Astra as orchestrator on High/Medium Astra is a big big model, which means it's capable of being really thorough and make really good executable plans → GPT-5.6 Luna as subagents/executors GPT-5.6 Luna Max is the same as GPT-6 Astra Light, while being 3.6x cheaper Once the execution is done, you let Astra review the code Repeat this in loop til you have it done!
This post quotes another post: x.com/thsottiaux/status/2098113585683808624