Plan guide

How much coding do you get from Claude Max 20×?

Available models: Fable 5.1 / Opus 5.5 / Sonnet 5.5. $200/mo buys about 90 h/week of building at p50 peak (p10–p90 50–180) (source: community). Coding-agent quality scores (/100): Fable 5.1 (max) (with fallback) · Claude Code · 62/100 planning, Sonnet 5.5 (high) · Claude Code · 55/100 building, Sonnet 5.5 (high) · Claude Code · 55/100 admin. Building model cost: $1.24 per benchmark task at API prices. Catalog note: Fable empty in 2–3 days is the planning number. Do not add Fable+Sonnet. Oct 2026: includes $200/month of Platform API credits, separate from these hours.

Hours by role

p10 / p50 / p90, h/week, at the plan's own clock. Peak and off-peak differ only where the vendor publishes a night rate.

RoleBenchmark quality /100Peak p10 / p50 / p90 (h/week)Off-peak p10 / p50 / p90 (h/week)
PlanningFable 5.1 (max) (with fallback) · Claude Code · 62/1003 / 8 / 15—
BuildingSonnet 5.5 (high) · Claude Code · 55/10050 / 90 / 180—
AdminSonnet 5.5 (high) · Claude Code · 55/10055 / 100 / 200—

Clock: no published off-peak discount. Runs as a CLI agent. Source: community.

Get Claude Max 20× →Rank it against your stack →

Current coding-agent evaluations

The ranking uses the fixed setting named in each role above. These scores do not establish subscription hours. Model availability verified 2026-11-09: vendor documentation.

Comparisons involving Claude Max 20×