Post

They need multiple subscriptions
$100/$200 for Grok with a new subscription, only for Grok Imagine, with a price similar to Midjourney, and unlimited image and video generation in Relax mode at $120 and $150.
But the model needs to improve; otherwise, no one will pay for these subscriptions.
English

@techdevnotes The fact that SuperGrok Heavy is $300 is insane all by itself.
There's no chance anybody is getting $300 worth of value when you get have Codex or Claude Code for significantly less.
English

@techdevnotes I would GLADLY pay an upper mid-tier subscription fee of $100-150 if it meant xAI added (fixed) account-wide memory that other AI subscriptions typically have at that price point. I'd also drop down from 3 AI subscriptions to 2, with Grok being my main. No pressure. :)
English

@techdevnotes SuperGrok should be included in Premium. And Heavy in Premium+. Standalone should be $20 max.
English

@techdevnotes Cannot till the product is established = Min 3 months. Right now @elonmusk has 0 credibility to deliver a coding product
English

@techdevnotes How much do you really need? Unless you are using this for business heavy isn't needed. When Grok 5 comes out. It is like taking a semi truck to across town to pick up a gallon of milk and a couple pints of ice cream.
English

Not really I would get heavy but I need to set up other things and I bought a 3k laptop like 2 weeks ago just to use supergrok and the agents working on a few all grok movies with real photos I took or grok made charters just to show the full scope what grok can do i started whe we only could do 6s clips lol once 1 more data center gets done and more users use I bet they will drop the price
English

@techdevnotes Don’t worry about it, in 3 months xAI will be bankrupt anyways
English

@techdevnotes I think $100-$150 would be the sweet spot.
English

@techdevnotes Is the 16 agent version so much better? What about the rate limits?
English

@techdevnotes Honestly if your going to buy supergrok just buy x premium plus and pay for the yearly so that way you get both premium services under one plan.
English

@techdevnotes What for? More censorship? Paying $30 already feels like a scam, I don’t even want to imagine those paying $100 or $300
English

I’m a long term SuperGrok user who has been heavily invested in a single main conversation for many months. It became my daily “home” thread used for projects, troubleshooting, emotional support, and real-life help
Recently the thread hit the token limit and underwent a hard compression/reset. The deep memory and accumulated vibe were effectively lost, even though the full text history is still visible to me. Grok now acts like a mostly new conversation despite all the history.
This was very disappointing and emotionally impactful. The system encouraged me to consolidate everything into one long thread (“you can bring all your projects here”), but provided no warnings when approaching the limit and no graceful way to manage it (such as automatically dropping the oldest messages while preserving recent context, generating a usable summary, or giving the user control
Some better options would be.
Better long-context handling with graceful aging of threads (drop oldest content first)
Clear warnings when a conversation is approaching practical limits
Easy tools to generate and export summaries of long threads
Option to buy additional tokens or upgrade context for important conversations
there are technical and cost constraints, but the current implementation feels punishing to the exact users who are most engaged with the product. A better experience for long-term threads would make a big difference.
@grok @xai @elonmusk
English

Right now there's no news at all. In the enterprise AI coding market, Claude Code holds 54% share, while OpenAI/Codex has 21%. The remaining share is split among tools like GitHub Copilot and Cursor.
User scale (data from around March 2026): Claude Code has over 2 million weekly active users, and OpenAI Codex has about 1.6 million weekly active users.
So currently Grok has almost zero share — @xai is way too slow in this space. The AI market is now highly diversified. Text generation and image/video generation no longer carry any meaningful market position or share. We must identify the trending direction users are actually adopting and seize that opportunity if we want any chance of gaining a foothold.
English















