Claude Code comes with Claude Pro at $20 a month ($17 billed yearly), Max from $100 and Team seats at $25 ($20 billed yearly). Across enterprise deployments, Anthropic puts the average at $150–250 per developer a month.
Claude CodeFable 5.1 (max) (with fallback)70.4 · Coding Agent Index
CodexGPT-6 Astra (max)67.0 · Coding Agent Index
Muse CodeMuse Spark 1.3 (xhigh)64.2 · Coding Agent Index
Grok BuildGrok 4.5 (high)64.1 · Coding Agent Index
Kimi Code CLIKimi K362.6 · Coding Agent Index
OpencodeGemini 3.8 Flash (high)61.2 · Coding Agent Index
Cursor CLIComposer 2.5 Fast38.3 · Coding Agent Index
Demo setups use current model families. The pairings above are the combinations benchmarked by Artificial Analysis; rankings measure performance, not popularity.
Illustrative daily trends · Other totals: last 30 days. Build Hours: this week, model-estimated activity including overlap; not engineer hours.
USED BY ENGINEERING TEAMS AT
How much does Claude Code cost?
Every paid Claude plan includes Claude Code, and it draws on the same usage pool as Claude on the web, desktop and mobile. Per person or seat, before tax:
Claude plans that include Claude Code, per person or seat. Prices verified October 5, 2026 against claude.com/pricing and Anthropic's Claude Code cost docs.
Plan
Price
Claude Code usage
Free
$0
Not included
Pro
$20 a month, or $17 billed annually
The base allowance; Fable only through usage credits
Max 5x
$100 a month
5x Pro's usage, with Fable inside the weekly limit
Max 20x
$200 a month
20x Pro's usage
Team Standard
$25 a seat a month, or $20 billed annually
More than Pro, for teams of 2 to 150
Team Premium
$125 a seat a month, or $100 billed annually
5x a Standard seat
Enterprise
$20 a seat a month, billed annually
All usage billed at API rates on top
Average spend
$150–250 per developer a month, about $13 per active day
Anthropic's figure across enterprise deployments; 90% of users stay under $30 a day
What does Claude Code cost a team of 10?
Seats are the floor. On Team, usage past a seat's allowance bills as usage credits at API rates, once an admin turns them on. On Enterprise, every token bills at API rates on top of the seat.
A team of 10 on Claude Code, per month before tax, at the plans table's prices (read October 5, 2026).
What the team buys
Arithmetic
Per month
Team Standard seats, billed annually
10 × $20
$200
Team Standard seats, billed monthly
10 × $25
$250
8 Standard and 2 Premium seats, billed annually
8 × $20 + 2 × $100
$360
Enterprise seats (20-seat minimum)
20 × $20, plus every token at API rates
From $400
Enterprise at Anthropic's average usage, 20 seats
20 × $20 + 10 × $150–250
$1,900–2,900
Is Claude Code free?
No. The Free plan doesn't include Claude Code; Pro is the cheapest plan that does. With an API key you pay only for the tokens you use, at the per-token rates below.
How much does Claude Code cost a heavy user?
A plan's price is the floor, not the cost. One engineer's month, recounted from the transcripts, came to more than a hundred times the subscription at API list prices, and would have cost 14 times more again without prompt caching:
One engineer's Claude Code month: 318 Claude Code sessions (September 4 to October 3, 2026), priced at Anthropic's API list prices (verified October 4, 2026 against claude.com/pricing). Not what was billed.
Measured
Value
Tokens
67.9 billion, 98.4% of them cache reads
Cost at API list price
$26,072
The same tokens without caching
$370,526
Where the cost went
Cache reads 57%, cache writes 33%, output 10%
Subagents' share
41% of tokens, 34% of cost
Median session
$49
Per merged pull request
$43 (601 merged)
Days with a usage-limit notice
16 of 29
Plan paid for
Claude Max 20x, plus usage credits
Claude Code API pricing per token
With an API key, and on Enterprise, Claude Code bills every token at these rates. Most of a long session is cache reads, at 0.05x input on Opus 5.5 and Sonnet 5.5, 0.025x on Fable 5.1 and 0.1x on Haiku 5.5. One request late in a session, on an API key (5-minute cache):
Opus 5.5, priced from the table below
One request
cache read 200,000 × $0.20 per million = $0.04
cache write 10,000 × $5 per million = $0.05
output 2,000 × $20 per million = $0.04
total $0.13
The same request without caching
input 210,000 × $4 per million = $0.84
output 2,000 × $20 per million = $0.04
total $0.88
Anthropic API prices per million tokens, for Claude Code on an API key or Enterprise. Prices verified October 11, 2026 against platform.claude.com/docs/en/about-claude/pricing.
Model
Input
Cache write (5 min / 1 hour)
Cache read
Output
Fable 5.1
$10
$12.50 / $20
$0.25
$50
Opus 5.5
$4
$5 / $8
$0.20
$20
Sonnet 5.5
$2
$2.50 / $4
$0.10
$10
Haiku 5.5 (prompts up to 100K tokens)
$0.10
$0.125 / $0.20
$0.01
$0.50
Claude Code usage limits by plan
Every plan's limits reset on a rolling five-hour window, and paid plans add weekly limits on top. Claude Code draws on the same pool as Claude on the web, desktop and mobile.
Anthropic publishes no fixed message count: what fits depends on the length of the conversation, the model and the features used. It gives ratios instead. Pro has at least 5x the usage of Free per five hours, Max 5x or 20x Pro, a Team Standard seat more than Pro and a Premium seat 5x Standard.
Fable runs on Pro only through usage credits. On Max and on Team Premium seats it can take up to half the weekly limit.
At the limit you wait for the reset, move to a bigger plan or turn on usage credits at standard API rates; Settings > Usage shows where you stand. On Team, owners set spend limits for the organization and for each member; Enterprise adds group limits.
How often they bite: a limit notice showed in 78 of that engineer's 318 sessions (25%), mostly the five-hour limit, used up 16 times; the weekly limit showed on 3 days. A status line can show both meters as you work.
Which plan fits: Pro, Max or a Team seat?
Pro fits occasional use. Anyone who runs agent sessions every day, or fans work out to subagents, needs Max: 5x for most, 20x for anyone who hits the five-hour limit on 5x most days.
On Team, Premium costs five times Standard for five times the usage: give it to the people who hit the limit; Claude Team and Enterprise plans compares the seats and what Enterprise bills. Zest prices each session it captures at list rates, inside the allowance too, so you can size each seat to the person (Claude Code usage).
What drives a team's Claude Code bill
Seats are fixed. Usage credits and per-token billing move with three things (Codex pricing and Cursor pricing work differently):
The model. Per token, Fable 5.1 costs 2.5 times Opus 5.5, and Opus 5.5 twice Sonnet 5.5; fast mode on Opus 5.5 costs twice the standard rate.
Cache lifetime. Claude Code keeps the conversation cached for an hour on a subscription within its plan, and for five minutes on usage credits or an API key unless promptCacheTtl is 1h. After a longer pause, the next request writes the whole context again.
Subagents. Each has its own cache, at a five-minute lifetime even on a subscription.
How to see Claude Code cost per engineer
Anthropic's own tools first:
Each developer: /usage prices the session at list price and, on a plan, splits the last 24 hours or 7 days by skill, subagent, plugin and MCP server.
Team: the spend report in org analytics shows token usage and estimated spend per user and model, as CSV. Usage inside the seat allowance isn't priced.
Enterprise: the Enterprise Analytics API returns per-user usage and cost; API organizations have the Console usage page and the Claude Code Analytics API.
Zest: the plugin prices every session at list rates, inside the allowance too, per engineer, and ties it to the pull request it led to.
/usage, Session block (example from Anthropic's cost docs)
Total cost: $0.55
Total duration (API): 6m 20s
Total duration (wall): 6h 33m 10s
Total code changes: 0 lines added, 0 lines removed
Usage by model:
claude-sonnet-4-6: 1.2k input, 5.3k output, 940.0k cache read, 50.0k cache write ($0.55)
TRY IT ON YOUR TEAM
See what your team ships with AI.
Starter is free (3 seats, no credit card). Pro is $49 a month with 5 seats and Max is $149 a month with 10 seats — priced per workspace, not per engineer, with extra seats at $19 a month.
Yes. Standard and Premium seats both include Claude Code, which draws on the seat's allowance alongside chat, and a Premium seat has 5x the usage of a Standard one.
What is Claude Max?+
Anthropic's individual plan for heavy use, with 5x or 20x the usage of Pro per five-hour session, billed monthly. Both sizes include Claude Code; the prices are in the plans table.
What are Claude usage credits?+
Pay-as-you-go usage past a plan's limits, billed at standard API rates, with an optional monthly spend limit; turn them on in Settings > Usage or with /usage-credits in Claude Code. They don't reset the five-hour window, and while you use them Claude Code caches the conversation for five minutes instead of an hour.
Winding / Teams
