Last verified: November 13, 2026 (Pacific) against claude.com/pricing, Ultra plan docs, and platform.claude.com pricing.
Prior note: Sources: Anthropic Help Center — Claude Code May–August 2026 weekly limits promotion (updated; permanent +25% now official) and Fable-on-plan article. Exact token counts are not published — check Settings → Usage or /usage in the CLI.
Direct Answer (November 13, 2026): Claude Code shares one account usage pool across CLI, IDE, desktop, and web with two meters on most plans: a rolling 5-hour session window and a weekly usage limit. On current published plans, Pro ($20/mo or $17/mo annual) gives at least 5× Free usage per 5-hour session; Max ($100/mo) provides 5× Pro usage every 5 hours and weekly; and Ultra ($200/mo, formerly Max 20×) provides 10× Pro weekly usage with no 5-hour session limit for verified accounts. Anthropic’s temporary May–September 13, 2026 +50% promo ended; current flagship Fable 5.2 ($8/$40 API) runs on usage credits on Pro and uses up to 50% of weekly limits on Max, Ultra, and Team Premium.
Claude Code Limits — Takeaway PDF
The two clocks
People treat “I hit my limit” as one event. It is two.
- 5-hour session. Rolling window. Burns when the model is working, not when you are idle. Wait/usage windows depend on your active plan (Pro gives at least 5× Free usage; Max gives 5× Pro usage; Ultra removes the 5-hour limit for verified accounts). Neither the +50% weekly promo nor the permanent +25% weekly change raised or lowered this clock.
- Weekly bucket. Fixed reset time on your account (Settings → Usage). This is the one that ends a Friday on Max after a Fable week. The May–September 2026 +50% promotion applied to Claude Code only — CLI, IDE, desktop, and web Code — not to regular Claude chat (Cowork has since been unified into core Claude). That temporary bump is over; check your live meter under Settings → Usage or
/usage.
Eligible for the promo and the permanent +25%: Pro, Max, Team, and legacy seat-based Enterprise. Not eligible: Free and consumption-based Enterprise.
The September 13–14 change & November 2026 Baseline
Historical promotion timeline (support.claude.com/en/articles/15910845):
- May 13 – Sep 13, 2026 11:59 PM PT: Temporary +50% weekly Claude Code promotion active for Pro, Max, Team, and seat-based Enterprise.
- Post-promotion published tiers (November 2026): Pro ($20/mo or $17/mo annual) gives at least 5× Free 5-hour session capacity; Max ($100/mo) gives 5× Pro usage every 5 hours and weekly (+ $100/mo API credits); Ultra ($200/mo, formerly Max 20×) gives 10× Pro weekly usage with no 5-hour limit for verified accounts (+ $200/mo API credits); Team Premium ($100/$125 per seat) gives 5× Standard seat usage.
Exact prompt/token quotas remain dynamic by task length, model choice, and attachment size — read your live meter in Settings → Usage.
Who gets Fable 5.2 on the subscription
Official rule on claude.com/pricing: current flagship Claude Fable 5.2 ($8/$40 API, superseding Fable 5.1 and Fable 5 at $10/$50) follows the shared Fable subscription access rules:
| Plan | Fable 5.2 on the subscription |
|---|---|
| Free ($0) | Not available |
| Pro ($20 / $17 annual) | Not in included allowance; bills via pay-as-you-go usage credits ($8/$40 MTok) from the first Fable token. |
| Team Standard ($20 / $25 seat) | Usage credits / pooled API credits ($8/$40 MTok). |
| Max ($100) / Ultra ($200, formerly Max 20×) | Included up to 50% of weekly limits (same pool as Sonnet and Opus). |
| Team Premium ($100 / $125 seat) | Included up to 50% of weekly limits (5× Standard seat usage). |
| Enterprise ($20/seat + API usage, 20-seat min) | Usage billed at standard API rates ($8/$40 for Fable 5.2). |
Note: Earlier one-time July 2026 migration credits have closed; Fable 5.2 runs at $8 input / $40 output ($0.20 cache read) on API or usage credits.
Practical read: Pro accesses Fable via usage credits; Max ($100/mo) and Ultra ($200/mo) include Fable up to half of weekly limits before overflowing to usage credits or Opus 5.5 ($4/$20) / Sonnet 5.5 ($2/$10).
What still burns quota
- Long agent loops and multi-agent fan-out (each worker keeps its own context).
- Dumping a whole repo when one file would do.
- MCP tool output that stays in context for the rest of the session.
- Leaving
ANTHROPIC_API_KEYset so the CLI bills API rates and ignores the subscription.
What to do now
- If you are on Pro and need Fable all week, budget credits or move the hard jobs to Max. The subscription does not include Fable on Pro.
- If you are on Max or Ultra, check Settings → Usage /
/usageagainst your current Max (5× Pro) or Ultra (10× Pro weekly, no 5-hour limit for verified accounts) limits, not the retired May–September +50% promotion. Route high-volume loops to Haiku 5.5 or Sonnet 5.5 and save Fable 5.2 for demanding reasoning tasks. - Do not upgrade from Pro to Max only because the 5-hour window feels tight on Friday. Check which clock you hit. Session vs week is a different purchase.
Keep This Guide: Claude Code Limits (PDF)
Part of the complete guide: Claude Code
Get The Signal
What we actually ran, what it measured, and what it means for a restoration shop — every week, including the flops.

Leave a Reply