Developers never really voiced any major desire to enter the age of AI coding; they always appeared quite happy to tap away at both chiclet and clacky mechanical keyboards into the wee small hours on manual coding tasks.
But that relaxed indifference changed once real coding assistants came onto the scene. Among the most appealing tools in this space is Anthropic’s Claude Code, an AI-powered command-line coding assistant that helps developers write, edit, debug and automate code.
But there’s no such thing as a free lunch or an unlimited Claude Code quota, even on the company’s $200 annual subscription deal.
Limited to a Relative Multiplier
According to Claude itself, “Max 20x ($200/month) is the top individual tier with 20x Pro usage, at which level rate limits stop being a practical concern for most full-day development work. That’s essentially the extent of the official promise, i.e., it’s a relative multiplier, not a hard number.”
Since March, 2026, Claude Code Max subscribers have been reporting quota exhaustion in as little as 19 minutes instead of the expected 5 hours.
A Reddit post from Anthropic states that, “We’re aware people are hitting usage limits in Claude Code way faster than expected. We’re actively investigating, and will share more when we have an update.”
When in Doubt, Try a Downgrade
While a sense of vagueness over exactly what each developer is entitled to and an equally nebulous explanation of the root cause of the problem from Anthropic, reports have suggested that there are bugs in the system that make token tracking inaccurate. Some developers have even reverted to downgrading their version of Claude Code to an older version in order to circumvent the current challenges.
Writing on X (formerly Twitter), Anthropic technical staff member Thariq Shihipar has noted that, “To manage growing demand for Claude, we’re adjusting our 5-hour session limits for free/Pro/Max subs during peak hours. Your weekly limits remain unchanged. During weekdays between 5 am–11 am PT / 1 pm–7 pm GMT, you’ll move through your 5-hour session limits faster than before.”
Cache-22
Some of the difficulties may also be associated with prompt caching, i.e., the software that Anthropic uses to optimize a developer’s API usage by allowing the user to “resume from specific prefixes” in their prompts. This significantly reduces processing time and costs for repetitive tasks or prompts with consistent elements.
Put simply, taking a break away from Claude Code could lead to higher costs when the developer decides to go back in.
Claude Code users won’t be delighted to learn that Claude itself suggests that all this “opacity is deliberate,” i.e., Anthropic frames limits in terms of sessions, resets and relative usage bands. There is no published “you get exactly X tokens” contract, which arguably (well, certainly, really) makes it pretty tough for software application development professionals to plan, schedule and budget their work effectively.
Hey guys, maybe it’s time to just go back to the old ways? We all know that’s not going to happen.
from DevOps.com https://ift.tt/WFAHvuJ
Comments
Post a Comment