This document provides an overview of the available Anthropic Claude models on Vertex AI, covering the following topics:
Available Claude models: Learn about the different Claude models, their capabilities, and their ideal use cases.
What's next: Find resources to start using the Claude models in your projects.
The Anthropic Claude models on Vertex AI are available as fully managed, serverless APIs. To use a Claude model on Vertex AI, you send a request directly to the Vertex AI API endpoint. Because the Anthropic Claude models use a managed API, you don't need to provision or manage infrastructure.
You can stream responses from Claude models to reduce perceived latency for end users. A streamed response uses server-sent events (SSE) to incrementally stream the response.
The following models are available from Anthropic to use in Vertex AI. To access a Claude model, go to its Model Garden model card.
Model
Description
Use Case
Claude Opus 4.1
Anthropic's most powerful model, which excels at coding and agent capabilities.
Complex, multi-step tasks; agentic search and analysis; expert-level coding.
Claude Opus 4
A powerful model for coding and agent capabilities.
Advanced coding; long-horizon tasks; AI agents; agentic search.
Claude Sonnet 4
Balances performance for coding with speed and cost for high-volume use cases.
Everyday development tasks; AI assistants; efficient research; large-scale content generation.
Claude 3.7 Sonnet
First Claude model to offer extended thinking for complex problem-solving.
Agentic coding; customer-facing agents; computer use; visual data extraction.
Claude 3.5 Sonnet v2
A powerful model for real-world software engineering and agentic capabilities.
Agentic tasks and tool use; complex coding; document Q&A; visual data extraction.
Claude 3.5 Haiku
Anthropic's fastest and most cost-effective model, with improvements across a range of skills.
Code completions; interactive chatbots; data extraction; real-time content moderation.
Claude 3 Haiku
Anthropic's fastest vision and text model for near-instant responses.
Live customer interactions; content moderation; cost-saving tasks; vision tasks.
Claude 3.5 Sonnet
Outperforms Claude 3 Opus on many evaluations with the speed and cost of Claude 3 Sonnet.
Coding; complex customer support queries; data science; visual processing.
Anthropic's Claude models support Vertex AI request-response logging. You can enable 30-day request-response logging of your prompt and completion activity to monitor usage and troubleshoot issues. For more information, see Log requests and responses.
Claude Opus 4.1
Claude Opus 4.1 is Anthropic's most powerful model, excelling at coding and agent capabilities, especially agentic search. It is well-suited for tasks that require advanced intelligence, such as the following:
AI agents: Enabling AI agents to complete complex, multi-step tasks with precision and reliability.
Agentic search and analysis: Connecting to multiple data sources to synthesize information and insights across different repositories.
Expert-level coding: Planning and executing complex coding tasks end-to-end, maintaining high-quality code that is consistent with your style.
Virtual collaboration: Using sustained reasoning capabilities to support use cases that involve long-horizon tasks and long chains of actions.
Content creation: Generating content with natural prose, including long-form content, technical documentation, marketing copy, and front-end design mockups.
Long context and memory: Incorporating memory capabilities that allow it to effectively summarize and reference previous interactions.
Claude Opus 4 is a powerful model for coding and agent capabilities, especially agentic search. It is well-suited for tasks that require advanced intelligence, such as the following:
Advanced coding: Independently planning and executing complex development tasks end-to-end. It can adapt to your style and maintain high code quality.
Long-horizon tasks and complex problem solving (virtual collaborator): Supporting use cases that involve long-horizon tasks that require memory, sustained reasoning, and long chains of actions.
AI agents: Enabling agents to tackle complex, multi-step tasks that require high accuracy.
Agentic search and research: Connecting to multiple data sources to synthesize comprehensive insights across repositories.
Content creation: Creating content with natural prose, including long-form creative content, technical documentation, marketing copy, and frontend design mockups.
Memory and context management: Incorporating memory capabilities that allow it to effectively summarize and reference previous interactions.
Claude Sonnet 4 balances performance for coding with speed and cost, making it suitable for high-volume use cases such as the following:
Coding: Handling everyday development tasks with enhanced performance, such as powering code reviews, bug fixes, API integrations, and feature development with immediate feedback loops.
AI Assistants: Powering production-ready assistants for real-time applications, from customer support automation to operational workflows that require both intelligence and speed.
Efficient research: Performing focused analysis across multiple data sources while maintaining fast response times. It is well-suited for rapid business intelligence, competitive analysis, and real-time decision support.
Large-scale content: Generating and analyzing content at scale with improved quality. You can create customer communications, analyze user feedback, and produce marketing materials with a balance of quality and throughput.
Claude 3.7 Sonnet is a highly capable model and the first Claude model to offer extended thinking—the ability to solve complex problems with step-by-step reasoning. With this model, you can balance speed and quality by choosing between standard thinking for near-instant responses or extended thinking for advanced reasoning.
For more information about extended thinking, see Anthropic's documentation.
Claude 3.7 Sonnet is optimized for the following use cases:
Agentic coding: This model excels at agentic coding and can complete tasks across the software development lifecycle, from initial planning to bug fixes, maintenance, and large refactors. It offers strong performance in both planning and solving for complex coding tasks, making it a good choice to power end-to-end software development processes.
Customer-facing agents: This model offers strong instruction following, tool selection, error correction, and advanced reasoning for customer-facing agents and complex AI workflows.
Computer use: This model is highly accurate for computer use, enabling you to direct Claude to use computer applications.
Content generation and analysis: This model excels at writing and can understand nuance and tone to generate high-quality content and perform deep content analysis.
Visual data extraction: With its strong vision skills, this model is a good choice for teams that want to extract raw data from visuals like charts or graphs as part of their AI workflow.
Claude 3.5 Sonnet v2 is a powerful model for real-world software engineering tasks and agentic capabilities. It delivers these advancements at the same price and speed as Claude 3.5 Sonnet.
The upgraded Claude 3.5 Sonnet model is capable of interacting with tools that can manipulate a computer desktop environment. For more information, see the Anthropic documentation.
This model is optimized for the following use cases:
Agentic tasks and tool use: Offers strong instruction following, tool selection, error correction, and advanced reasoning for agentic workflows that require tool use.
Coding: For software development tasks ranging from code migrations, code fixes, and translations, this model offers strong performance in both planning and solving for complex coding tasks.
Document Q&A: Combines strong context comprehension, advanced reasoning, and synthesis to deliver accurate and conversational responses.
Visual data extraction: With its strong vision skills, this model can extract raw data from visuals like charts or graphs as part of AI workflows.
Content generation and analysis: Can understand nuance and tone in content, generating high-quality content and performing deep content analysis.
Claude 3.5 Haiku, the next generation of Anthropic's fastest and most cost-effective model, is well-suited for use cases where speed and affordability are important. It improves on its predecessor across a range of skills. This model is optimized for the following use cases:
Code completions: With its rapid response time and understanding of programming patterns, this model excels at providing quick, accurate code suggestions and completions in real-time development workflows.
Interactive chatbots: This model's improved reasoning and natural conversation abilities make it well-suited for creating responsive, engaging chatbots that can handle high volumes of user interactions efficiently.
Data extraction and labeling: Leveraging its improved analysis skills, this model 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 content understanding, this model provides fast, reliable content moderation for platforms that require immediate response times at scale.
Anthropic's Claude 3 Haiku is Anthropic's fastest vision and text model for near-instant responses to basic queries. It is designed for AI experiences that mimic human interactions. It's optimized for the following use cases:
Live customer interactions and translations.
Content moderation to catch suspicious behavior or customer requests.
Cost-saving tasks, such as inventory management and knowledge extraction from unstructured data.
Vision tasks, such as processing images to return text output, and analyzing charts, graphs, technical diagrams, reports, and other visual content.
Anthropic's Claude 3.5 Sonnet outperforms Claude 3 Opus on a wide range of Anthropic's evaluations, with the speed and cost of Anthropic's mid-tier Claude 3 Sonnet model. This model is optimized for the following use cases:
Coding: Writing, editing, and running code with sophisticated reasoning and troubleshooting capabilities.
Customer support: Handling complex queries by understanding user context and orchestrating multi-step workflows.
Data science and analysis: Navigating unstructured data and leveraging multiple tools to generate insights.
Visual processing: Interpreting charts and graphs that require visual understanding.
Content writing: Writing content with a more natural, conversational tone.
