[FEATURE] Transparent, predictable usage limits for Max plan ($200/month) professional users

Resolved 💬 3 comments Opened Mar 7, 2026 by chapa254 Closed Mar 10, 2026

Preflight Checklist

  • [x] I have searched existing requests and this feature hasn't been requested yet
  • [x] This is a single feature request (not multiple features)

Problem Statement

I'm on the Max 20x plan — the most expensive individual tier available. I use Claude Code professionally as my primary development tool, and it's also the core of my business infrastructure.

This week my token budget started depleting significantly faster than usual with the exact same usage patterns. Nothing changed on my end.

The real impact: I bill clients on a weekly basis and need to estimate how much work I can deliver within my weekly token window. Right now I have no way to do that because the limits are opaque and change without notice. I literally cannot price my services reliably. And when I hit my weekly limit with 2-3 days still to go, the "extra usage" pricing is obscenely expensive — it feels like being punished for actually using the tool I'm already paying top dollar for.

The core issue isn't speed or priority — it's predictability. Lowering my request priority after hitting a limit? Totally fine. I can wait. But silently raising the effective token cost per interaction makes it impossible to run a business on this tool.

I genuinely love Claude. It's the best in the market and I've built my entire workflow around it. But I need to be able to rely on it to make money with it. Right now I can't.

Proposed Solution

  • Published, concrete limits per plan. Not "dynamic." Not "fair use." Actual numbers.
  • A usage dashboard showing real-time token consumption so users can plan their workday.
  • Advance communication when limits change. A status page update, an email, anything. Silent degradation is unacceptable at $200/month.
  • Separate priority from cost. Throttle speed during peak demand if needed — that's fine. But don't change how many tokens each interaction costs without telling users.
  • Graceful degradation instead of hard stops. When I hit my limit, don't cut me off completely. Decrease performance gradually — slower responses, lower priority, reduced reasoning — but let me keep working. A hard stop in the middle of a workday is a dealbreaker for professional users.
  • Reasonable overage pricing. If I'm already on the most expensive plan, hitting the weekly limit shouldn't force me to either stop working or pay through the nose just to finish the week.

Alternative Solutions

_No response_

Priority

High - Significant impact on productivity

Feature Category

Configuration and settings

Use Case Example

_No response_

Additional Context

_No response_

View original on GitHub ↗

This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗