AI Coding Assistants
Anthropic’s coding agent for terminal, IDE, web, desktop, mobile, and SDK workflows
Pricing: From $17/mo + usage. Best for Developers delegating multi-file implementation, debugging, refactoring, and test work inside a real repository and Engineering teams that want the same coding agent across terminal, IDE, desktop, web, mobile, and Slack surfaces.
Overview
Claude Code is Anthropic’s coding-agent product for working directly with a codebase. It can inspect repositories, edit files, run commands, and coordinate development work from the terminal, supported IDEs, desktop, web, mobile, Slack, and programmatic Agent SDK surfaces. Opus 5 is a model choice inside Claude Code, not a separate coding product; its availability and default status depend on the account, provider, organization policy, and installed Claude Code version.
Claude Code works best for developers and engineering teams that want an agent to handle multi-file changes, debugging, refactoring, testing, and longer implementation tasks while keeping the repository and tool loop in view. The core caveat is commercial complexity: subscription sign-in shares capacity with Claude, usage credits can add metered overflow, Console authentication meters tokens separately, and Fast mode has its own premium rate.
Start with an eligible Claude subscription for human-steered interactive work. Choose a Team or Enterprise workspace when seat controls and governance matter, and use Console or API-funded routes for unattended automation, shared budgets, or workloads that need explicit per-token controls.
Pricing
Pricing routes
App subscriptions, API usage, team workspaces, and enterprise contracts can follow different billing paths. See the pricing guide for current plan and meter details.
From $17/mo
Claude Code is included with eligible paid Claude plans. Pro is the lowest paid individual entry at $17 per month with annual billing ($200 upfront) or $20 month to month; usage is shared with Claude.
Bundled access
Eligible individual, Team, and seat-based Enterprise users can continue beyond included limits with separately purchased usage credits. Standard-mode overflow is billed at standard API rates and remains subject to spend controls.
Usage-based API
Console-authenticated Claude Code is metered separately from app subscriptions. Standard Opus 5 pricing is $5 per million input tokens and $25 per million output tokens.
Usage-based API
Fast mode is a research-preview Opus configuration billed from the first Fast-mode token at $10 per million input tokens and $50 per million output tokens through Claude Console or subscription usage credits.
From $20/seat/mo
Every current Team seat includes Claude Code. Standard seats start at $20 per seat per month with annual billing, while Premium seats provide a larger included usage allowance.
From $20/seat/mo
Current self-serve Enterprise starts at $20 per seat plus usage at API rates; sales-assisted and older seat-based arrangements can have different contract, allowance, and usage-credit boundaries.
Usage-based API
Usage: Anthropic API: Claude Opus 5 costs $5 per million input tokens and $25 per million output tokens at standard speed. Anthropic recommends starting with Opus 5 for complex agentic coding and enterprise work; standard pricing is separate from fast mode and Claude subscription quota.
Usage-based API
Usage: Anthropic API: Claude Sonnet 5 is available for Claude Code and platform workflows. Introductory pricing is $2 per 1M input tokens and $10 per 1M output tokens through August 31, 2026, then $3 per 1M input and $15 per 1M output tokens.
Usage-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.
Input types
text, image, code, file, url
Output types
text, code, file
Supported platforms
Web, iOS, Android, Mac, Windows, Linux
Delivery modes
Web app, Mobile app, Desktop app, Chat app, CLI, IDE extension, SDK
Company
Anthropic
Founded
Not specified
Headquarters
Not specified
Category
AI Coding Assistants
Evidence boundary
Editorial guidance grounded in official product sources.
FAQ
No. Claude Code is Anthropic’s coding-agent product, while Opus 5 is one model that Claude Code can run. On the Anthropic API, the opus alias resolves to Opus 5, and Claude Code version 2.1.219 or later is required.
Claude Code is included with Pro, Max, current Team seats, and current Enterprise seats when the user authenticates with the eligible Claude account. The Free Claude plan does not list Claude Code. Older Enterprise contracts can use different seat definitions.
Anthropic documents Opus 5 as the default for Max, Team Premium, usage-based Enterprise, and Anthropic API accounts. Pro, Team Standard, and subscription-seat Enterprise accounts default to Sonnet 5. Provider routing, organization defaults, allowlists, and explicit user settings can change the active model.
No. Anthropic documents Opus with 1M context as included on Max, Team, and Enterprise, while Pro requires usage credits; Anthropic API and pay-as-you-go routes have full 1M access. The 128K maximum is an Opus 5 synchronous Messages API specification and should not be treated as a guaranteed output limit for Claude app or every Claude Code product surface.
No. Fast mode is a research-preview configuration for Opus 5 and Opus 4.8. Subscription users need usage credits, and billing begins with the first Fast-mode token at $10 per million input tokens and $50 per million output tokens. It is not supported in the VS Code extension or on the listed third-party cloud routes.
No. Claude subscriptions and Claude Console or API billing are separate. Claude Code can use the subscription allocation when signed in with a Claude account, but an ANTHROPIC_API_KEY takes precedence and can move the session to separately billed API usage.
Recent updates
Event date
Version 2.1.241
Claude Code 2.1.241 delivered bug fixes and reliability improvements.
Official source Changelog - Claude Code Docs
Verified
Event date
Version 2.1.240
Claude Code 2.1.240 delivered bug fixes and reliability improvements.
Official source Changelog - Claude Code Docs
Verified
Event date
Version 2.1.239
Claude Code 2.1.239 updated cost estimates, added a fullscreen renderer offer, added claude-api upgrade support, and improved synced plugins and platform reliability.
Official source Changelog - Claude Code Docs
Verified
Event date
Version 2.1.238
Claude Code 2.1.238 added readline-style keybindings and command-generated headers for plugin marketplace and catalog fetches, alongside reliability fixes.
Official source Changelog - Claude Code Docs
Verified
Related reading
AI Coding Assistants
AI code editor with agents, context-aware completion, Bugbot, and cloud workflows.
Last verified August 24, 2026
Read profile
AI Coding Assistants
GitHub-native AI coding assistant for chat, code review, and agent workflows.
Last verified August 16, 2026
Read profile
AI Coding Assistants
OpenAI's coding agent with GPT-5.6 model choice, selectable reasoning, and parallel Ultra workflows.
Last verified July 27, 2026
Read profile
AI Coding Assistants
Windsurf is now Devin Desktop: an agentic IDE for local and cloud coding workflows.
Last verified August 13, 2026
Read profileDecision rail
Keep the key facts, primary action, and section jumps visible while you evaluate the profile.
Last verified August 16, 2026
On this page
Share
Pass this page along
Copy the link or send it to the channel where your team compares tools, pricing, and tradeoffs.
Navigation
Open related pages for pricing, reviews, and adjacent coverage.
Open direct comparison pages involving this tool.
Nearby tools in the same lane.
Reference links