The Anthropic Claude models on Vertex AI offer fully managed andserverless models as APIs. To use a Claude model on Vertex AI, senda request directly to the Vertex AI API endpoint. Because the AnthropicClaude models use a managed API, there's no need to provision or manageinfrastructure.

You can stream your Claude responses to reduce the end-user latency perception.A streamed response uses server-sent events (SSE) to incrementally stream theresponse.

You pay for Claude models as you use them (pay as you go), or you pay a fixedfee when usingprovisionedthroughput. Forpay-as-you-go pricing, seeAnthropic Claude models on the Vertex AIpricing page.

Available Claude models

Anthropic's Claude models support Vertex AI request-responselogging. Enable 30-day request-response logging of your prompt and completionactivity to track any model misuse by your users. For more information, seeLogrequests and responses.

Claude Opus 4.6

Claude Opus 4.6 is the next generation of Anthropic's most intelligentmodel, and the world's best model for coding, enterprise agents, andprofessional work.

Retirement Date: Not sooner than February 5, 2027.

• Long-running agents: Power production-ready assistants for multi-step,real-time applications—from customer support automation to complex operationalworkflows that require peak accuracy, intelligence, and speed.
• Coding: Handle everyday development tasks with enhanced performance––orplan and execute complex software projects spanning hours or days––with theability to save, maintain, and reference information across multiple sessions.
• Cybersecurity: Deploy agents that autonomously patch vulnerabilitiesbefore exploitation––shifting from reactive detection to proactive defense.
• Financial analysis: Conduct entry-level financial analysis, deliveradvanced predictive analysis, or preemptively develop intelligent riskmanagement strategies that leverage best-in-class domain knowledge.
• Computer use: Claude Opus 4.6 is our most accurate model forcomputer use, enabling developers to direct Claude to use computers the waypeople do.
• Business tasks: Generate and edit office files like slides, documents, andspreadsheets with minimal input.
• Research: Perform focused analysis across multiple data sources, turningexpert analysis into final deliverables. Ideal for complex problem solving,rapid business intelligence, and real-time decision support.

Go to the Claude Opus 4.6 model card

Claude Opus 4.5

Claude Opus 4.5 is an industry leader across coding, agents, computeruse, and enterprise workflows.

Retirement Date: Not sooner than Nov 24, 2026.

• Agents: Claude Opus 4.5, paired with our advanced tool usecapabilities, enables more capable agents with new behaviors.
• Coding: Opus 4.5 can confidently deliver multi-day software developmentprojects in hours, working independently with the technical depth and tasteto create efficient and straightforward solutions. It has improvedperformance across coding languages, with better planning and architecturechoices—making it the ideal model for enterprise developers.
• Enterprise workflows: Opus 4.5 can power agents that manage sprawlingprofessional projects from start to finish. It better leverages memory tomaintain context and consistency across files, alongside a step-changeimprovement in creating spreadsheets, slides, and docs.
• Financial analysis: Opus 4.5 connects the dots across complexinformation systems—regulatory filings, market reports, internal data—makingsophisticated predictive modeling and proactive compliance possible.
• Cybersecurity: Opus 4.5 brings professional-grade analysis to securityworkflows, correlating logs, vulnerability databases, and threatintelligence for proactive threat detection and automated incident response.
• Computer use: Our best computer-using model yet,Claude Opus 4.5 navigates new experiences with confident,consistent approaches that deliver more human-like browsing, enabling betterweb QA, workflow automation, and advanced user experiences.

Go to the Claude Opus 4.5 model card

Claude Sonnet 4.6

Claude Sonnet 4.6 delivers frontier intelligence at scale—built for coding, agents, and enterprise workflows.

Retirement Date: Not sooner than Feb. 17th 2027.

• Long-running agents: Power production-ready assistants for multi-step,real-time applications—from customer support automation to complex operationalworkflows that require peak accuracy, intelligence, and speed.
• Coding: Handle everyday development tasks with enhanced performance––orplan and execute complex software projects spanning hours or days––with theability to save, maintain, and reference information across multiple sessions.
• Cybersecurity: Deploy agents that autonomously patch vulnerabilitiesbefore exploitation––shifting from reactive detection to proactive defense.
• Financial analysis: Conduct entry-level financial analysis, deliveradvanced predictive analysis, or preemptively develop intelligent riskmanagement strategies that leverage best-in-class domain knowledge.
• Computer use: Claude Sonnet 4.6 is our most accurate model forcomputer use, enabling developers to direct Claude to use computers the waypeople do.
• Business tasks: Generate and edit office files like slides, documents, andspreadsheets with minimal input.
• Research: Perform focused analysis across multiple data sources, turningexpert analysis into final deliverables. Ideal for complex problem solving,rapid business intelligence, and real-time decision support.

Go to the Claude Sonnet 4.6 model card

Claude Sonnet 4.5

Claude Sonnet 4.5 is Anthropic's Sonnet-class model for poweringreal-world agents, with industry leading capabilities around coding, computeruse, cybersecurity, and working with office files like spreadsheets.

Retirement Date: Not sooner than Sept 29, 2026.

• Long-running agents: Power production-ready assistants for multi-step,real-time applications, from customer support automation to complexoperational workflows that require peak accuracy, intelligence, and speed.
• Coding: Handle everyday development tasks with enhanced performance - orplan and execute complex software projects spanning hours or days - with theability to save, maintain, and reference information across multiplesessions.
• Cybersecurity: Deploy agents that autonomously patch vulnerabilitiesbefore exploitation, shifting from reactive detection to proactive defense.
• Financial analysis: Conduct entry-level financial analysis, deliveradvanced predictive analysis, or preemptively develop intelligent riskmanagement strategies that leverage best-in-class domain knowledge.
• Computer use: Anthropic's most accurate model for computeruse, enabling developers to direct the model to use computers the way peopledo.
• Business tasks: Generate and edit office files like slides, documents,and spreadsheets with minimal input.
• Research: Perform focused analysis across multiple data sources, turningexpert analysis into final deliverables. Ideal for complex problem solving,rapid business intelligence, and real-time decision support.

Go to the Claude Sonnet 4.5 model card

Claude Opus 4.1

Claude Opus 4.1 is Anthropic's Opus-class model and an industryleader for coding and agent capabilities, especially agentic search. It excelsfor customers needing frontier intelligence:

Retirement Date: Not sooner than Aug 5, 2026.

• AI agents: Enable AI agents to complete complex, multi-step tasks withprecision and reliability.
• Agentic search and analysis: Connect to multiple data sources tosynthesize information and insights across different repositories.
• Expert-level coding: Plan and execute complex coding tasks end-to-end,maintaining high-quality code that is consistent with your style.
• Virtual collaboration: Use the sustained reasoning capabilities tounlock new use cases involving long-horizon tasks and long chains ofactions.
• Content creation: Generate content with human-quality, natural prose.Create long-form content, technical documentation, marketing copy, andfront-end design mockups.
• Long context and memory: Incorporates memory capabilities that allow itto effectively summarize and reference previous interactions.

Go to the Claude Opus 4.1 model card

Claude Haiku 4.5

Claude Haiku 4.5 delivers near-frontier performance for a wide range ofuse cases, and stands out as one of the best coding models in the world—with theright speed and cost to power free products and high-volume user experiences.

Retirement Date: Not sooner than Oct 15, 2026.

• Free tier user experiences: Claude Haiku 4.5 deliversnear-frontier performance at a cost and speed that makes free agent productsand agentic use cases economically viable at scale.
• Latency-sensitive experiences: Claude Haiku 4.5's speed isideal for real-time applications like customer service agents and chatbotswhere response time is critical.
• Coding sub-agents: Use Claude Haiku 4.5 to power sub-agents,enabling multi-agent systems that tackle complex refactors, migrations, andlarge feature builds with quality and speed.
• Financial analysis: Use Claude Haiku 4.5 to monitor thousandsof data streams—tracking regulatory changes, market signals, and portfoliorisks to preemptively adapt compliance and trading systems at previouslyimpossible scales.
• Research sub-agents: Perform parallel analyses across multiple datasources while maintaining fast response times. Ideal for rapid businessintelligence, competitive analysis, and real-time decision support.
• Business tasks: Claude Haiku 4.5 is capable of producing andediting office files like slides, documents, and spreadsheets. It alsobetter supports strategy and campaign planning, business analysis andbrainstorming.

Go to the Claude Haiku 4.5 model card

Claude Opus 4

Claude Opus 4 is a state-of-the-art model for coding and agentcapabilities, especially agentic search. It excels for customers needingfrontier intelligence:

Retirement Date: Not sooner than May 14, 2026.

• Advanced coding: Independently plan and execute complex developmenttasks end-to-end. It adapts to your style and maintains high code qualitythroughout.
• Long-horizon tasks and complex problem solving (virtual collaborator):Unlock new use cases that involves long-horizon tasks that require memory,sustained reasoning, and long chains of actions.
• AI agents: Enable agents to tackle complex, multi-step tasks thatrequire peak accuracy.
• Agentic search and research: Connect to multiple data sources tosynthesize comprehensive insights across repositories.
• Content creation: Create human-quality content with natural prose.Produce long-form creative content, technical documentation, marketing copy,and frontend design mockups.
• Memory and context management: Incorporates memory capabilities thatallow it to effectively summarize and reference previous interactions.

Go to the Claude Opus 4 model card

Claude Sonnet 4

Claude Sonnet 4 balances impressive performance for coding with theright speed and cost for high-volume use cases:

Retirement Date: Not sooner than May 14, 2026.

• Coding: Handle everyday development tasks with enhancedperformance—power code reviews, bug fixes, API integrations, and featuredevelopment with immediate feedback loops.
• AI Assistants: Power production-ready assistants for real-timeapplications—from customer support automation to operational workflows thatrequire both intelligence and speed.
• Efficient research: Perform focused analysis across multiple datasources while maintaining fast response times. Ideal for rapid businessintelligence, competitive analysis, and real-time decision support.
• Large-scale content: Generate and analyze content at scale with improvedquality. Create customer communications, analyze user feedback, and producemarketing materials with the right balance of quality and throughput.

Go to the Claude Sonnet 4 model card

Claude 3.5 Haiku

Claude 3.5 Haiku is optimized for use cases where speed andaffordability matter. It improves on its predecessor across every skill set.Claude 3.5 Haiku is optimized for the following use cases:

• Code completions - With its rapid response time and understanding ofprogramming patterns, Claude 3.5 Haiku excels at providingquick, accurate code suggestions and completions in real-time developmentworkflows.
• Interactive chat bots - Claude 3.5 Haiku's improved reasoningand natural conversation abilities make it ideal for creating responsive,engaging chatbots that can handle high volumes of user interactionsefficiently.
• Data extraction and labeling - Leveraging its improved analysis skills,Claude 3.5 Haiku efficiently processes and categorizes data,making it useful for rapid data extraction and automated labeling tasks.
• Real-time content moderation - With strong reasoning skills and contentunderstanding, Claude 3.5 Haiku provides fast, reliablecontent moderation for platforms that require immediate response times atscale.

Go to the Claude 3.5 Haiku model card

Claude 3 Haiku

Anthropic's Claude 3 Haiku is Anthropic's fastest vision and text model fornear-instant responses to basic queries, meant for seamless AI experiencesmimicking human interactions.

• Live customer interactions and translations.

• Content moderation to catch suspicious behavior or customer requests.

• Cost-saving tasks, such as inventory management and knowledge extractionfrom unstructured data.

• Vision tasks, such as processing images to return text output, analysis ofcharts, graphs, technical diagrams, reports, and other visual content.

Go to the Claude 3 Haiku model card

What's next

Learn how to use Anthropic's models.