Users slam Claude Code opaque limits: same workload now eats weekly quota in 22 hours vs 4.4 days
sujingshen · x · 2026-10-23
A developer's open letter to Anthropic argues Claude Code's usage limits have become opaque and possibly silently tightened. Many users abandoned $200 Codex plans for Claude Code because it felt unlimited, only to hit weekly caps in 3 days doing identical work. Key datapoint: token volume that took 4.4 days the prior week pushed the weekly meter from 0% to 54% in 22 hours. By contrast, when Codex users raised similar complaints, OpenAI responded with numbers (40,000 accounts on a wrong meter), while Anthropic shipped only a warning, and its pricing page reserves the right to limit usage "at our discretion." Commenter sujingshen notes the real issue isn't price but unpredictability — showing a transparent daily token meter would beat another explainer post.
Related event: Anthropic investigating reports of unusually fast Claude usage burn(2 posts)→
More from Models
- Unverified chart claims GPT-6 Sol tops DeepSecBench Pareto frontier at 80% lower cost — JohnPhamous · 2026-10-23
- Developer: LLMs that constantly push back make work feel like arguing all day — louisvarge · 2026-10-23
- Gemini 4 Argon spotted testing apps autonomously, and 'topping humor benchmarks' — andrew_n_carr · 2026-10-23
- Gemini 4 reportedly far less sensitive in refusals than other models — haider1 · 2026-10-23
- SemiAnalysis: Anthropic subscriptions offer 5x+ more value than OpenAI as $200 plan value halves — RexDouglass · 2026-10-23
- Google Cloud's Physical-Engineering Last Exam: all frontier models score under 50% — Afinetheorem · 2026-10-23