Claude Sonnet 4.6
Claude Sonnet 4.6
Version: 1
AnthropicLast updated May 2026
Claude Sonnet 4.6 delivers frontier intelligence at scale—built for coding, agents, and enterprise workflows. With a 1M token context window and 128K max output, Sonnet 4.6 is ideal for coding, agents, office tasks, financial analysis, cybersecurity, and c
Reasoning

Models from Partners and Community

These models constitute the vast majority of the Azure AI Foundry Models and are provided by trusted third-party organizations, partners, research labs, and community contributors. These models offer specialized and diverse AI capabilities, covering a wide array of scenarios, industries, and innovations. An example of models from Partners and community are the family of large language models developed by Anthropic. Anthropic includes Claude family of state-of-the-art large language models that support text and image input, text output, multilingual capabilities, and vision. See Anthropic's privacy policy to know more about privacy. Learn how to deploy Anthropic models . Characteristics of Models from Partners and Community:
  • Developed and supported by external partners and community contributors.
  • Diverse range of specialized models catering to niche or broad use cases.
  • Typically validated by providers themselves, with integration guidelines provided by Azure.
  • Community-driven innovation and rapid availability of cutting-edge models.
  • Standard Azure AI integration, with support and maintenance managed by the respective providers.
Models from Partners and Community are deployable as Managed Compute or serverless API deployment options. The model provider selects how the models are deployable.

Key capabilities

About this model

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

Key model capabilities

  • Image & text input: With strong vision capabilities, Claude Sonnet 4.6 can process images and return text outputs to analyze and understand charts, graphs, technical diagrams, reports, and other visual assets.
  • Computer use: Sonnet 4.6 delivers confident, consistent navigation with more human-like browsing, enabling developers to direct Claude to use computers the way people do.
  • Advanced tool use: Build agents that can take action with the tool search tool, programmatic tool calling, and tool use examples.
  • Adaptive thinking: Adaptive thinking is an upgrade to extended thinking that gives Claude the freedom to think as much or as little as needed depending on the task and effort level.

Use cases

See Responsible AI for additional consideration for responsible use.

Key use cases

Claude Sonnet 4.6 delivers frontier intelligence at scale—built for coding, agents, and enterprise workflows. Use cases include:
  • Agents: Sonnet 4.6 excels at complex, multi-step tasks requiring sustained reasoning and adaptive decision-making—ideal for workflows where reliability and autonomy matter most.
  • Coding: Sonnet 4.6 is built for iterative development work, handling complex codebases without losing quality as you guide it through building, refactoring, and debugging. It can compress multi-day projects into hours with the technical depth to deliver production-ready solutions.
  • Enterprise workflows: Sonnet 4.6 powers agents that manage professional projects from start to finish, leveraging memory to maintain context across files with a step-change improvement in creating spreadsheets, slides, and docs.
  • Financial analysis: Sonnet 4.6 connects the dots across regulatory filings, market reports, and internal data—enabling sophisticated modeling and proactive compliance.
  • Cybersecurity: Sonnet 4.6 brings professional-grade analysis to security workflows, correlating logs, vulnerability databases, and threat intelligence for proactive threat detection and automated incident response.
  • Computer use: Sonnet 4.6 delivers confident, consistent navigation with more human-like browsing—enabling better web QA, workflow automation, and advanced user experiences.

Out of scope use cases

Please refer to the Claude Sonnet 4.6 system card .

Pricing

Pricing is based on a number of factors. See pricing details here . Azure credits cannot be applied to Claude models. Usage will be billed to the credit card associated with your subscription. Users are advised to review documentation here for applicable pricing and terms prior to deployment to avoid unintended charges.

Technical specs

Please refer to the Claude Sonnet 4.6 system card .

Training cut-off date

August 2025

Input formats

Image & text input: With powerful vision capabilities, Claude Sonnet 4.6 can process images and return text outputs to analyze and understand charts, graphs, technical diagrams, reports, and other visual assets. Text output: Claude Sonnet 4.6 can output text of a variety of types and formats, such as prose, lists, Markdown tables, JSON, HTML, code in various programming languages, and more.

Supported language

Claude Sonnet 4.6 can understand and output a wide variety of languages, such as English, French, Standard Arabic, Mandarin Chinese, Japanese, Korean, Spanish, and Hindi. Performance will vary based on how well-resourced the language is.

Supported Azure regions

Global

Sample JSON response

Input

{
  "model": "claude-sonnet-4-6",
  "max_tokens": 1024,
  "messages": [
    {
      "role": "user",
      "content": "Hi, who are you?"
    }
  ]
}

Output

200: {
  "content": [
    {
      "text": "Hi! My name is Claude.",
      "type": "text"
    }
  ],
  "id": "msg_01XFDUDYJgAACzvnptvVoYEL",
  "model": "claude-sonnet-4-6",
  "role": "assistant",
  "stop_reason": "end_turn",
  "stop_sequence": null,
  "type": "message",
  "usage": {
    "input_tokens": 31,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "cache_creation": {
      "ephemeral_5m_input_tokens": 0,
      "ephemeral_1h_input_tokens": 0
    },
    "output_tokens": 25,
    "service_tier": "standard"
  }
}
4XX: {
  "error": {
    "message": "Invalid request",
    "type": "invalid_request_error"
  },
  "request_id": "",
  "type": "error"
}

Model architecture

Please refer to the Claude Sonnet 4.6 system card .

Long context

Claude Sonnet 4.6 has a 1M token context window.

Optimizing model performance

Please refer to the Claude Sonnet 4.6 system card .

Additional assets

  • Claude Documentation : Visit Claude documentation for a wealth of resources on model capabilities, prompting techniques, use case guidelines, and more.
  • Adaptive Thinking Guide : Understand how best to use extended thinking with Claude.
  • Claude Prompting Resources : Check out Anthropic's prompting tools and guides to learn how to craft prompts that elicit more helpful, nuanced responses.
  • Claude Cookbooks : Check out example code for a variety of complex tasks, such as RAG from various web sources, making SQL queries, function calling, multimodal prompting, and more.

Distribution channels

  • Claude API: For developers interested in building agents, Claude Sonnet 4.6 is available on the Claude Developer Platform.
  • Claude Code: Use Claude Sonnet 4.6 with Anthropic's industry-leading coding agent, Claude Code.

More information

N/A

Safety techniques

The Claude Sonnet 4.6 system card describes in detail the wide range of evaluations Anthropic ran to assess the model's safety and alignment.

Safety evaluations

Claude Sonnet 4.6 represents incremental improvements over Claude Sonnet 4.5, with enhancements in reasoning quality and overall performance. The Claude Sonnet 4.6 system card details comprehensive safety evaluations including safeguards testing, agentic safety evaluations for computer use and coding capabilities, assessments of reward hacking behavior, Usage Policy compliance testing, alignment evaluations covering a range of misalignment risks, and model welfare assessments. It also includes capability evaluations related to the production of certain harmful technologies, in line with our Responsible Scaling Policy commitments.

Known limitations

Please refer to the Claude Sonnet 4.6 system card .

Acceptable use

Acceptable use policy

Anthropic's Usage Policy is intended to help users stay safe and promote the responsible use of Anthropic products and services.

Terms of Service

Terms of Service Link

Claude is a proprietary model developed by Anthropic. Usage is governed by Anthropic's Commercial Terms of Service for API access.

Quality and performance evaluations

Quality and performance evaluations

BenchmarkTest nameClaude Sonnet 4.6 score
Agentic terminal codingTerminal-Bench 2.059.1%
Agentic codingSWE-bench Verified79.6%
Agentic computer useOSWorld Verified72.5%
Agentic tool useτ2-benchTelecom: 97.9%, Retail: 91.7%
Scaled tool useMCP Atlas61.3%
Agentic searchBrowseComp74.7%
Multidisciplinary reasoningHumanity’s Last ExamWithout tools: 33.2%, With tools: 49.0%
Agentic financial analysisFinance Agent v1.163.3%
Office tasksGDPVal-AA Elo1633
Novel problem solvingARC AGI 258.3%
Graduate-level reasoningGPQA Diamond89.9%
Visual reasoningMMMU ProWithout tools: 74.5%, With tools: 75.6%
Multilingual Q&AMMMLU89.3%

Benchmarking methodology

Please refer to the Claude Sonnet 4.6 system card for benchmarking methodology.

Public data summary

N/A
Model Specifications
Context Length1000000
Quality Index0.74
Training DataAugust 2025
Last UpdatedMay 2026
Input TypeText,Image,Code
Output TypeText,Image,Code
ProviderAnthropic
Languages8 Languages
Related Models