@thsottiaux I don’t like the idea of burning through tokens even faster and then having to wait 5 hours to finish my work. The only acceptable option would be to double both the 5-hour and weekly limits. Then 2x speed would make sense.
Important note regarding Grok @Bot:
Going forward, @SpaceX will use the best back end model for any given task, including Claude Opus 5.5, MidJourney, Suno and other leading APIs.
Whatever is most likely to give you the best outcome.
Claude Cloud isn’t worth it at all. I ran the exact same task in Cloud and locally, and the token usage was massively higher in Cloud.
Locally: 1% of the 5-hour limit
Cloud: 22% of the 5-hour limit
This is on the 5x plan. I wouldn’t recommend Claude Cloud unless you just want to burn through your tokens.
@thsottiaux When you call, it doesn’t answer.
The ability to install Docker.
When starting a local chat, Dot sometimes asks you to enter permissions there because it can’t resolve it by itself. @thsottiaux
Found a limitation with ChatGPT Dot: it can’t install Docker by itself, which is a pretty big downside right now. I’d love to see this fixed in a future update @thsottiaux@OpenAI