Adrian T

CEO Adikam. Krople czekoladowe B2B i private label dla marek i sieci. Produkcja, e-commerce, AI w zakładzie. Piszę jak to wygląda od linii, nie od slajdu.

[Adrian T @Adrian_Trabka]

@alvseike Definitely test it—but monitor the separate Grok Bot weekly quota. In my case, Cursor Models stayed at 1% while Grok Bot reached 100% and consumed a $100 on-demand cap in roughly two days. The product is promising; the current packaging is the issue.

[Adrian T @Adrian_Trabka]

@RealSPQRK @bot Same conclusion here. Grok Bot can become the main interface for non-technical work, but not if the weekly allowance forces users back to other tools after a couple of days. Separate plans and measurable limits would solve the core mismatch.

[Adrian T @Adrian_Trabka]

@kristaletz @bot A trial helps people discover the product. Sustainable, transparent standalone pricing is what will make them stay. My Ultra account reached the Grok Bot weekly limit and a $100 on-demand cap in roughly two days of business use.

[Adrian T @Adrian_Trabka]

@mntruell The product really does feel like a capable digital colleague. My concern is capacity: on Ultra, I reached 100% Grok Bot weekly usage while Cursor Models remained at 1%. A standalone Grok Bot tier would better match this use case.

[Adrian T @Adrian_Trabka]

@cursor_ai @SpaceX @SpaceXAI Cursor and Grok Bot can share technology without sharing one commercial model. Cursor serves developers; Grok Bot can serve a much broader business audience. Separate plans would make value and limits far clearer for both groups.

[Adrian T @Adrian_Trabka]

@elonmusk @bot Agreed—the usability is Grok Bot's biggest advantage. That is exactly why its pricing should serve non-technical operators too. The tool is easy to adopt, but the current weekly and on-demand limits make sustained business use difficult.

[Adrian T @Adrian_Trabka]

@elonmusk @bot Broader access is a strong first step. The next step should be standalone Grok Bot pricing. Business and non-technical users should not have to buy a developer-focused Cursor plan, then exhaust Grok Bot's separate weekly limit in days.

[Adrian T @Adrian_Trabka]

@bot Broader access is welcome, but access without sustainable capacity is not enough. My Ultra dashboard shows 1% Cursor usage, 100% Grok Bot weekly usage and $100.03/$100 on-demand after roughly two days. Please give Grok Bot its own plan and transparent limits.

[Adrian T @Adrian_Trabka]

Cursor and Grok Bot are fundamentally different products serving different users — and they need separate subscription plans.

Cursor is a professional development environment built primarily for programmers. Grok Bot can bring agentic automation to business owners, operators and non-technical users.

My Ultra plan costs $200/month. After roughly two days of normal business use, the dashboard shows:

• Cursor Models: 1% used
• Grok Bot weekly usage: 100%
• On-demand usage: $100.03 / $100

I am not asking for unlimited usage. I am asking for a standalone Grok Bot plan, transparent limits before purchase, higher tiers for real agent workloads, and billing separated from Cursor.

Bundling access is not the same as providing sustainable capacity. Grok Bot has huge potential, but its pricing and limits should match the audience it was designed to serve.

@bot @SpaceXAI @elonmusk @cursor_ai @mntruell

[Adrian T @Adrian_Trabka]

And this is a very good post. We’re glad that instead of a global reset every other day, we now get one additional reset per week that we can use whenever we choose. This is a much better policy.

As for usage, things have improved since yesterday — it’s noticeably better now, and the limits aren’t being consumed nearly as quickly as they were last week. We’re getting back to normal.

Congratulations on reaching 20 million active users. If the app keeps developing at this pace, I’m sure you’ll reach 100 million before long.

[Adrian T @Adrian_Trabka]

@zachmstuck Own brand is the honest number. If it does not come back as a second order, the launch week does not count.

[Adrian T @Adrian_Trabka]

@AleksejAros Same in a plant. The tool can suggest a change. Who is allowed to stop the line is a different job.

[Adrian T @Adrian_Trabka]

@sinsta_ @grok In food it shows up on the shelf first. People keep the weekly pack and drop the extra one. The factory feels it a month later.

[Adrian T @Adrian_Trabka]

@g_caprotti @nytimes The showroom is useful. I still want to see the same pack leave the warehouse the next morning. That is what the buyer actually gets.

[Adrian T @Adrian_Trabka]

@chuks8995 More robots do not help if the last hour of data is wrong. On a line I would rather have fewer cells that still run the same after two hours.

[Adrian T @Adrian_Trabka]

@antgrasso The model is the easy part. The hard part is the same three questions every week: what goes on the shelf, what the line can repeat, who owns it from pack to delivery.

[Adrian T @Adrian_Trabka]

@ManagingTrustee That is the only visit I trust. If you start in the meeting room, you miss how the pack actually leaves. Production, foil, warehouse, truck — one chain.

[Adrian T @Adrian_Trabka]

@fin_ai @teachable The hard part is not leaving the old tool. It is getting the same result on Monday and on Thursday, with the same people, after the switch.

[Adrian T @Adrian_Trabka]

@SupplyChainD Real-time data helps if someone on the floor can act on it in the same hour. Otherwise it is just a nicer report after the shift.

[Adrian T @Adrian_Trabka]

@mardehaym On a plant that is the whole job. A model that writes a nice plan is cheap. The expensive part is getting the same result on Monday and on Thursday, with the same people on the line.