Access & Pricing

PUBLIC

Access channels

CLI

Access via command-line interface.

IDE extension

Plugin for an IDE (VS Code, JetBrains, etc.).

Desktop app

Native app for macOS, Windows or Linux.

Web app

Access via web browser with no installation required.

Mobile app

Native app for iOS or Android.

SDK

Developer library for embedding the AI system in other applications.

API

Programmatic access via REST or WebSocket API.

Connected apps / integrations

Integration with external apps via standard connectors or MCP.

Availability scope

Global

GlobalNorth AmericaEuropean UnionUnited KingdomJapanAustralia & New ZealandIndiaKorea

Claude Code operates globally in countries supported by Anthropic (Supported Countries and Regions list). Key markets: US, Canada, UK, EU (all member states), Japan, South Korea, Australia, New Zealand, India, most of Latin America. NOT available in mainland China, Russia, Belarus, Iran, Cuba, Syria, North Korea, or other OFAC-sanctioned regions. EU data residency is available exclusively on Enterprise plans (Anthropic API EU region + Bedrock Frankfurt/Ireland + Vertex AI europe-west1/europe-west4).

Subscription model

PaidBusiness subscriptionUsage-based

Pricing

Updated 18 Jun 2026Official pricing

Claude Code is included in Claude subscription plans: Pro (USD 17/mo with annual billing, USD 20/mo monthly), Max 5x (USD 100/mo), Max 20x (USD 200/mo), and in Team and Enterprise premium seats. With an Anthropic Console account, Claude Code consumes API tokens at standard API pricing. Fast mode for Opus 4.8 on consumption-based plans: USD 30 / USD 150 per 1M tokens (input/output).

CategoryPlanUnitInputCachedOutput

Pro

Per month with annual subscription (USD 200 billed up front). USD 20 if billed monthly. Claude Code included.

standardflat$17——

Max 5x

Per month. Claude Code included, higher usage limits.

standardflat$100——

Max 20x

Per month. Claude Code included, highest access to Claude models.

standardflat$200——

Text tokens

High-speed Opus 4.8 configuration (2.5x faster). Research preview, available on consumption-based plans / via usage credits.

usage based1M tok.$30—$150

Security & Enterprise

Security source

Audit Logs

Audit logs for all admin actions (invites, permission changes, seat mgmt) available in Claude Console on Team and Enterprise plans. 12-month retention.

Data Residency

EU data residency on Enterprise: Anthropic API EU region, AWS Bedrock (Frankfurt eu-central-1, Ireland eu-west-1), Google Vertex AI (europe-west1, europe-west4).

Enterprise Access

SSO/SAML 2.0 + SCIM 2.0 provisioning, role-based access control (Owner/Admin/Member), domain capture, per-user seat management, custom OAuth apps.

Compliance Controls

SOC 2 Type II, ISO 27001, ISO 42001 (AI management), HIPAA BAA (Enterprise), GDPR, CCPA. Trust Center: trust.anthropic.com.

Encryption Controls

TLS 1.3 in transit; AES-256 at rest. All API endpoints HTTPS-only. User secrets (API keys) stored in OS keychain (macOS Keychain, Windows Credential Manager, libsecret).

CMEK

Customer-Managed Encryption Keys available via AWS Bedrock (AWS KMS) and Google Vertex AI (Cloud KMS) when using Claude Code with these backends.

Private Networking

PrivateLink (AWS Bedrock) and Private Service Connect (Google Vertex AI) for Claude Code traffic without traversing the public internet. Direct Anthropic API does not offer PrivateLink.

Context Caching

Prompt caching (up to 90% reduction in input token cost, up to 85% latency reduction) natively available for Claude Sonnet 4.5 and Opus 4 - Claude Code automatically caches CLAUDE.md, skill definitions, and conversation history.

Claude Code runs locally in the terminal and talks directly to the model APIs, with no backend server or remote code index. It asks permission before changing files or running commands. Team and Enterprise plans (premium seats) and Claude Code Enterprise are available.

Notes

Upd. 18 Jun 2026

Claude Code runs locally in the user's terminal and communicates directly with the Anthropic API (Claude API), Amazon Bedrock, or Google Vertex AI - with no intermediary backends or remote code index. Zero Data Retention (ZDR): default 30-day retention of prompts/responses for all plans; Enterprise customers can negotiate ZDR with retention limited to 24h (for abuse detection only). Training on data: user code is NOT used to train Anthropic models by default (opt-in only). Tool permissions: every shell command execution, file edit, or MCP call requires user consent (permission prompt) - --allowlist / --deny-list modes exist along with auto-approve for whitelisted commands. Authentication: OAuth 2.0 (Claude Console) or API key. SSO/SAML + SCIM available only on Team/Enterprise plans. Compliance: SOC 2 Type II, ISO 27001, ISO 42001, HIPAA BAA (Enterprise), GDPR-ready, CCPA. Sub-agents run in isolated subprocesses with their own context window - they have no access to the main context or session.

AnthropicActiveFeatured

Claude Code

Name: Claude Code
Brand: Anthropic

Claude Code is an agentic coding tool created by Anthropic. It reads your entire codebase, edits files, runs commands, and integrates with your development tools. It works in the terminal, IDEs (VS Code, Cursor, JetBrains), a desktop app, the browser, and Slack — all surfaces use the same underlying engine, so CLAUDE.md files, settings, and MCP servers work consistently everywhere.

History

Claude Code was announced on February 24, 2025 as a limited research preview, alongside the Claude 3.7 Sonnet model. It reached general availability in May 2025 with Claude 4. On October 20, 2025, Anthropic launched a web version of Claude Code and a sandboxing feature. The tool gained significant popularity during the 2025-2026 winter holidays.

Capabilities

Agentic search — understands project structure and dependencies without manual context-file selection.
Coordinated multi-file edits, bug fixing, writing and running tests, refactoring.
Git workflow — stages changes, writes commit messages, creates branches and pull requests. In CI: GitHub Actions, GitLab CI/CD.
Computer use (since March 2026) — opens apps, navigates the browser, runs dev tools.
Agent teams and dynamic workflows — execution across tens to hundreds of parallel sub-agents with result verification. Agent SDK for building custom agents.
Customization via CLAUDE.md, skills, hooks, and integration with external data sources through MCP (Model Context Protocol).

Security and control

Claude Code runs locally in your terminal and talks directly to the model APIs, with no backend server or remote code index. It asks for permission before making changes to files or running commands. It works on macOS, Linux, and Windows.

CLIIDE extensionDesktop appWeb appMobile appSDKAPIConnected apps / integrationsPaidBusiness subscriptionUsage-based

Released 24 Feb 2025Updated 15 Jul 2026Global · Global · North America · European Union · United Kingdom · Japan · Australia & New Zealand · India · Korea

Visit website API documentation