API boundary
Claude subscription access and Anthropic API or Console billing are separate systems. API-key usage is metered separately and can override subscription authentication when ANTHROPIC_API_KEY is set.
Pricing
Claude Code Pricing: Claude Plan Access and API Costs
Claude Code pricing depends on the route: Claude subscription access, team workspace rollout, or separate Anthropic API and Console billing.
AI Coding Assistants
Pricing checked May 26, 2026
Buyer guide
Where to start before you compare routes
Use the route summary and access paths first, then treat the API pricing section as the direct fact layer for metered developer spend.
Tracks
Which route fits whom
Claude subscription route
Use Claude plan access when you want first-party Claude Code usage tied to Claude subscription limits.
Best for: Solo developers and teams already buying Claude Pro, Max, Team, or Enterprise.
Avoid if: You need API keys, separate platform billing, or programmatic cost controls.
API and Console route
Use Anthropic API pricing when Claude Code usage should be metered, reported, and capped in the developer Console.
Best for: Automation builders and teams that need usage-based billing or workspace spend limits.
Avoid if: You only need human-driven Claude Code usage inside a Claude subscription.
Team rollout route
Use a pilot group, Console reporting, and spend limits before rolling Claude Code out broadly.
Best for: Managers modeling per-developer active-day and monthly usage ranges.
Avoid if: You need a simple public subscription comparison table.
Access paths
Subscription, API, and workspace routes
Use this section to separate what is bundled with Claude Code from routes that need a different pricing page, meter, or sales conversation.
Bundled appShared subscription quotaRecommended route
Claude plan access
Use Claude Code through an eligible Claude subscription when the work is human-driven and should follow Claude plan access and usage limits.
Best for: Individual developers and teams already choosing a Claude subscription tier.
Boundary: This route follows Claude plan access. It is not prepaid Anthropic API token spend.
Open Claude pricing contextDirect APISeparate API meter
Anthropic API and Console
Use API or Console billing when Claude Code usage should be tracked and capped through Anthropic developer-platform billing.
Best for: Platform teams, automation, and organizations that need API-key billing, Console reporting, or workspace spend limits.
Boundary: API-key usage is metered separately and can override subscription authentication when ANTHROPIC_API_KEY is set.
Open Claude pricing contextTeam workspaceIncluded in subscription
Team and enterprise workspace
Use the workspace route when admin controls, usage reporting, rate limits, and spend governance matter as much as plan access.
Best for: Teams rolling Claude Code out across multiple developers or enterprise environments.
Boundary: Workspace governance does not remove the need to separate Claude subscription access from API-token billing.
Open Claude Code pricing contextDirect API pricing
Direct API pricing
Use these metered API facts for developer budgets. App subscriptions and bundled routes stay in the route cards above.
Plans listed
3
Benchmark plan
Claude API (Opus 4.7)
API track
API plans
3 plans
Claude API (Opus 4.7)
APIUsage-based API
Usage: Anthropic API: Claude Opus 4.7 costs $5 per 1M input tokens and $25 per 1M output tokens; use for higher-capability coding-agent work when the extra spend is justified.
- Token-metered Anthropic API billing
- Full 1M token context window at standard pricing
- Separate from Claude subscription quota
Claude API (Sonnet 4.6)
APIUsage-based API
Usage: Anthropic API: Claude Sonnet 4.6 costs $3 per 1M input tokens and $15 per 1M output tokens; Claude Code API and Console billing should be modeled from token usage.
- Token-metered Anthropic API billing
- Tracked through Anthropic Console usage reporting
- Separate from Claude subscription quota
Claude API (Haiku 4.5)
APIUsage-based API
Usage: Anthropic API: Claude Haiku 4.5 costs $1 per 1M input tokens and $5 per 1M output tokens; use for lower-cost supporting automation where it fits the task.
- Token-metered Anthropic API billing
- Lower-cost API option for supporting automation
- Separate from Claude subscription quota
Free plan
No
Trial
No trial listed
Billing unit
Hybrid
Pricing checked
May 26, 2026
Watchouts
What buyers often miss
These are the boundary conditions and purchase traps worth checking before you optimize for the lowest headline number.
Claude prices belong on Claude pricing
Claude Code can be reached through Claude subscriptions, but this page should not duplicate the full Pro, Max, Team, or Enterprise plan table.
API keys can change the billing route
If ANTHROPIC_API_KEY is set, Claude Code prioritizes API-key authentication and charges API pay-as-you-go rates even if the user also has a Claude subscription.
Usage estimates are not fixed subscription prices
Claude Code cost varies by model, codebase size, number of running instances, and automation. Treat per-developer cost ranges as planning inputs, not fixed plan prices.
Editorial pricing notes
Pricing notes
Official plan caveats, contract details, and feature access notes that do not fit into the summary cards above.
Buying path
Start by deciding whether Claude Code is being used through Claude plan access or through Anthropic API and Console billing. The subscription route fits developers working interactively with Claude Code, while the API route fits teams that need programmatic control, metering, or production automation.
Upgrade triggers
Upgrade when coding sessions become frequent enough that lower access limits or lighter models slow the workflow. The signal is not one difficult task; it is repeated use across repositories, refactors, reviews, or agentic coding work where capacity and model choice affect delivery.
API and team boundaries
Treat subscription authentication, API-key billing, and team workspace access as separate budget lanes. API-key usage can change the billing owner and metering model, while team rollouts need policy, onboarding, and repository governance decisions beyond a single developer subscription.
Final pricing check
Before committing, verify which authentication path the team will actually use, who owns API keys, and whether spend should be forecast as seats or metered usage. If the workflow is still experimental, keep the route narrow until coding volume and ownership are clear.
Decision archive
Price history snapshots
Track how Claude Code pricing has moved over time, including plan lineup shifts, free access changes, and starting price updates.
1 archived snapshot
LatestPaid · Usage based
Last confirmed
May 26, 2026
First archived April 14, 2026
Latest archived pricing state remains unchanged since it was first recorded.
View source pageStarting price
Not specified
Access model
Paid access only
Plan count
3
Billing unit
Usage based
Claude API (Opus 4.7)
api-opus-4-7
Monthly: Not listed
Annual: Not listed
Usage: Anthropic API: Claude Opus 4.7 costs $5 per 1M input tokens and $25 per 1M output tokens; use for higher-capability coding-agent work when the extra spend is justified.
Claude API (Sonnet 4.6)
api-sonnet-4-6
Monthly: Not listed
Annual: Not listed
Usage: Anthropic API: Claude Sonnet 4.6 costs $3 per 1M input tokens and $15 per 1M output tokens; Claude Code API and Console billing should be modeled from token usage.
Claude API (Haiku 4.5)
api-haiku-4-5
Monthly: Not listed
Annual: Not listed
Usage: Anthropic API: Claude Haiku 4.5 costs $1 per 1M input tokens and $5 per 1M output tokens; use for lower-cost supporting automation where it fits the task.
FAQ
Claude Code pricing FAQ
Is Claude Code priced as a separate subscription?
Claude Code pricing depends on the access route. Interactive access can come through eligible Claude plans, while API-key and Console workflows should be budgeted from Anthropic API token pricing.
Does Claude Code API usage share my Claude app subscription?
No. Anthropic API usage is metered separately from Claude app subscriptions, so developer builds should use the API pricing lane rather than assuming subscription allowance covers API calls.
Which Claude API model should I use for Claude Code budgets?
Use Opus when the task needs the highest coding capability, Sonnet as the mainstream balance, and Haiku for lower-cost supporting automation where the model is sufficient.
How should teams think about Claude Code costs?
Teams should separate Claude plan access, workspace governance, API token spend, and usage reporting. The right budget depends on whether work happens in the app experience or through API/Console workflows.
Internal links
What to open next
Keep researching Claude Code
Pair the pricing snapshot with verdict, alternatives, and the full profile page.
ToolProfile: Claude CodeAnthropic's agentic coding assistant for terminal, IDE, browser, and automation workflows.Review: Claude Code ReviewClaude Code is one of the strongest agentic coding tools for deep repo work, but its CLI-first workflow and hybrid pricing make it a better fit for serious developers than casual users.Alternatives: Best Claude Code AlternativesClaude Code is excellent for terminal-first agentic development, but Cursor, GitHub Copilot, and Codex can be better fits if you care more about editor UX, GitHub workflows, or OpenAI-native cloud agents.UpdatesClaude Code changelogRecent product updates, fixes, and feature releases.
Head-to-head pages
Open direct comparison pages before choosing a plan.
VSClaude Code vs GitHub CopilotClaude Code is the better default for developers and engineering teams that want an agent-first coding workflow for scoped implementation tasks. GitHub Copilot is the better fit when GitHub-native rollout, broad IDE coverage, pull request assistance, and enterprise seat governance matter more than terminal-agent depth.VSClaude Code vs WindsurfClaude Code is the stronger long-term choice for teams that care about rollout friction, admin controls, and mixed-surface execution. Windsurf remains the better pick for buyers who want one AI-native IDE with previews and tighter in-editor guidance.VSClaude Code vs CodexThis comparison is a real tie: Claude Code is stronger for local-first coding workflows and deployment flexibility, while Codex is stronger for cloud orchestration and supervisory multi-agent work.VSClaude Code vs CursorClaude Code is the better overall pick because its wins land in the highest-leverage parts of engineering work: repo-wide execution, automation, and approval-first control in your own environment. Cursor remains excellent if you want the strongest AI editor, broader model choice, and richer built-in collaboration surfaces.
Other reviews in this category
Sanity-check nearby tools before committing to a pricing tier.
ReviewTabnine ReviewTabnine is strongest for security-conscious teams that need private AI coding across IDEs, terminal workflows, and controlled deployments.ReviewCodex Review: Is OpenAI's GPT-5.5 Coding Agent Worth It?Codex is easier to recommend because GPT-5.5 brings stronger long-horizon coding, a 400K context window, and better token efficiency than GPT-5.4 for most users, but the product still asks buyers to navigate subscriptions, credits, and separate API billing.ReviewGitHub Copilot ReviewGitHub Copilot remains one of the easiest AI coding assistants to adopt, combining GitHub-native workflow fit, broad IDE coverage, code review, and flexible plans from free to enterprise.ReviewCursor ReviewCursor is one of the strongest AI coding environments available, pairing deep codebase context with agentic workflows, but its usage-based pricing makes plan selection less predictable.