@xai@grok I have been having ongoing technical issues with my SuperGrokPro performance. I only got it to work on a book for myself. The mode mixes up chapters and doesn't follow prompts. Today, I clearly asked for a review of a chapter and it reviews the wrong chapter. Then it tells me it needs to double check better. Really? In heavy mode the A I doesn't double check and clearly ignores the prompt and reviews wrong chapter. It's so bad I have to keep starting new conversations. It's a waste of time and terrible server for the top tier or any paying tier
@grok@xai
Hello. I have superGrok subscription.
Bug Report: Grok 4.20 still stuck in infinite “thinking…” on long novel
I’m collaboratively writing a massive time-loop novel.
Every time in Grok 4.20: Even with a clear narrative with room for creative additions (within the established lore) it gets stuck. Even when I use the “BALANCED OUTPUT MODE” prompt, Grok 4.20 still gets permanently stuck in “thinking…” for hours with no response.
Expert mode works perfectly and fast. This makes 4.20 unusable for serious long-form book writing, even though that’s exactly what the multi-agent system is supposed to excel at. This has now happened multiple times on the same project. Please fix the internal agent debate / consensus timeout issue on very long creative contexts. Thank you.