Loading...
    • Developer Guide
    • API Reference
    • MCP
    • Resources
    • Release Notes
    Search...
    ⌘K
    First steps
    Intro to ClaudeQuickstart
    Models & pricing
    Models overviewChoosing a modelWhat's new in Claude 4.5Migrating to Claude 4.5Model deprecationsPricing
    Build with Claude
    Features overviewUsing the Messages APIContext windowsPrompting best practices
    Capabilities
    Prompt cachingContext editingExtended thinkingEffortStreaming MessagesBatch processingCitationsMultilingual supportToken countingEmbeddingsVisionPDF supportFiles APISearch resultsStructured outputs
    Tools
    OverviewHow to implement tool useFine-grained tool streamingBash toolCode execution toolProgrammatic tool callingComputer use toolText editor toolWeb fetch toolWeb search toolMemory toolTool search tool
    Agent Skills
    OverviewQuickstartBest practicesUsing Skills with the API
    Agent SDK
    OverviewQuickstartTypeScript SDKTypeScript V2 (preview)Python SDKMigration Guide
    MCP in the API
    MCP connectorRemote MCP servers
    Claude on 3rd-party platforms
    Amazon BedrockMicrosoft FoundryVertex AI
    Prompt engineering
    OverviewPrompt generatorUse prompt templatesPrompt improverBe clear and directUse examples (multishot prompting)Let Claude think (CoT)Use XML tagsGive Claude a role (system prompts)Prefill Claude's responseChain complex promptsLong context tipsExtended thinking tips
    Test & evaluate
    Define success criteriaDevelop test casesUsing the Evaluation ToolReducing latency
    Strengthen guardrails
    Reduce hallucinationsIncrease output consistencyMitigate jailbreaksStreaming refusalsReduce prompt leakKeep Claude in character
    Administration and monitoring
    Admin API overviewWorkspacesUsage and Cost APIClaude Code Analytics API
    Console
    Log in
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...
    Loading...

    Solutions

    • AI agents
    • Code modernization
    • Coding
    • Customer support
    • Education
    • Financial services
    • Government
    • Life sciences

    Partners

    • Amazon Bedrock
    • Google Cloud's Vertex AI

    Learn

    • Blog
    • Catalog
    • Courses
    • Use cases
    • Connectors
    • Customer stories
    • Engineering at Anthropic
    • Events
    • Powered by Claude
    • Service partners
    • Startups program

    Company

    • Anthropic
    • Careers
    • Economic Futures
    • Research
    • News
    • Responsible Scaling Policy
    • Security and compliance
    • Transparency

    Learn

    • Blog
    • Catalog
    • Courses
    • Use cases
    • Connectors
    • Customer stories
    • Engineering at Anthropic
    • Events
    • Powered by Claude
    • Service partners
    • Startups program

    Help and security

    • Availability
    • Status
    • Support
    • Discord

    Terms and policies

    • Privacy policy
    • Responsible disclosure policy
    • Terms of service: Commercial
    • Terms of service: Consumer
    • Usage policy
    Documentation

    Building with Claude

    Claude is a family of highly performant and intelligent AI models built by Anthropic. While Claude is powerful and extensible, it's also the most trustworthy and reliable AI available. It follows critical protocols, makes fewer mistakes, and is resistant to jailbreaks—allowing enterprise customers to build the safest AI-powered applications at scale.
    • What you can do with Claude
    • Enterprise considerations
    • Implementing Claude
    • Start building with Claude

    This guide introduces Claude's enterprise capabilities, the end-to-end flow for developing with Claude, and how to start building.

    What you can do with Claude

    Claude is designed to empower enterprises at scale with strong performance across benchmark evaluations for reasoning, math, coding, and fluency in English and non-English languages.

    Here's a non-exhaustive list of Claude's capabilities and common uses.

    CapabilityEnables you to...
    Text and code generation
    • Adhere to brand voice for excellent customer-facing experiences such as copywriting and chatbots
    • Create production-level code and operate (in-line code generation, debugging, and conversational querying) within complex codebases
    • Build automatic translation features between languages
    • Conduct complex financial forecasts
    • Support legal use cases that require high-quality technical analysis, long context windows for processing detailed documents, and fast outputs
    Vision
    • Process and analyze visual input, such as extracting insights from charts and graphs
    • Generate code from images with code snippets or templates based on diagrams
    • Describe an image for a user with low vision
    Tool use
    • Interact with external client-side tools and functions, allowing Claude to reason, plan, and execute actions by generating structured outputs through API calls

    Enterprise considerations

    Along with an extensive set of features, tools, and capabilities, Claude is also built to be secure, trustworthy, and scalable for wide-reaching enterprise needs.

    FeatureDescription
    Secure
    • Enterprise-grade security and data handling for API
    • SOC II Type 2 certified, HIPAA compliance options for API
    • Accessible through AWS (GA) and GCP (in private preview)
    Trustworthy
    • Resistant to jailbreaks and misuse. We continuously monitor prompts and outputs for harmful, malicious use cases that violate our AUP.
    • Copyright indemnity protections for paid commercial services
    • Uniquely positioned to serve high trust industries that process large volumes of sensitive user data
    Capable
    • 200K token context window for expanded use cases, with future support for 1M
    • Tool use, also known as function calling, which allows seamless integration of Claude into specialized applications and custom workflows
    • Multimodal input capabilities with text output, allowing you to upload images (such as tables, graphs, and photos) along with text prompts for richer context and complex use cases
    • Developer Console with Workbench and prompt generation tool for easier, more powerful prompting and experimentation
    • SDKs and APIs to expedite and enhance development
    Reliable
    • Very low hallucination rates
    • Accurate over long documents
    Global

    Implementing Claude

    Start building with Claude

    When you're ready, start building with Claude:

    • Follow the Quickstart to make your first API call
    • Check out the API Reference
    • Explore the Prompt Library for example prompts
    • Experiment and start building with the Workbench
    • Check out the Claude Cookbook for working code examples
    • Great for coding tasks and fluency in English and non-English languages like Spanish and Japanese
    • Enables use cases like translation services and broader global utility
    Cost conscious
    • Family of models balances cost, performance, and intelligence
    1. 1

      Scope your use case

      • Identify a problem to solve or tasks to automate with Claude.
      • Define requirements: features, performance, and cost.
    2. 2

      Design your integration

      • Select Claude's capabilities (e.g., vision, tool use) and models (Opus, Sonnet, Haiku) based on needs.
      • Choose a deployment method, such as the Claude API, AWS Bedrock, or Vertex AI.
    3. 3

      Prepare your data

      • Identify and clean relevant data (databases, code repos, knowledge bases) for Claude's context.
    4. 4

      Develop your prompts

      • Use Workbench to create evals, draft prompts, and iteratively refine based on test results.
      • Deploy polished prompts and monitor real-world performance for further refinement.
    5. 5

      Implement Claude

      • Set up your environment, integrate Claude with your systems (APIs, databases, UIs), and define human-in-the-loop requirements.
    6. 6

      Test your system

      • Conduct red teaming for potential misuse and A/B test improvements.
    7. 7

      Deploy to production

      • Once your application runs smoothly end-to-end, deploy to production.
    8. 8

      Monitor and improve

      • Monitor performance and effectiveness to make ongoing improvements.