Microsoft Foundry Adds Claude Sonnet 4.6: Near-Opus Performance at Lower Cost

Microsoft Foundry has expanded its AI model lineup with the addition of Claude Sonnet 4.6, a powerful new model that brings near-Opus-level performance to enterprise developers at a more accessible price point. The model is now available to teams building at scale across coding, agents, and professional workflows.

Frontier Performance with Cost Efficiency

Claude Sonnet 4.6 represents a significant upgrade from its predecessor, delivering intelligence that approaches Claude Opus 4.6 capabilities while maintaining better token efficiency than Claude Sonnet 4.5. This positions Sonnet 4.6 as an ideal choice for teams requiring high-performance AI without the premium pricing of top-tier models.

The model introduces several key capabilities that make it particularly suited for enterprise deployment:

Extended Context and Adaptive Reasoning

Sonnet 4.6 features a 1 million token context window (currently in beta), matching the extended context capabilities of Claude Opus 4.6. This massive context allows teams to work across entire codebases, lengthy financial models, and multi-document analysis without fragmentation or repeated context resets.

The model also introduces adaptive thinking and effort parameters, giving Claude the autonomy to determine when reasoning is required. This represents an evolution from traditional extended thinking approaches, optimizing the balance between performance and speed. Teams can fine-tune quality-latency-cost tradeoffs using these parameters.

Enhanced Computer Use Capabilities

With a 72.5% score on OSWorld Verified, Claude Sonnet 4.6 is Anthropic's most capable computer use model to date. The model demonstrates improved precision in clicking accuracy on difficult UI elements, enabling browser automation at scale without API key dependency.

This capability allows Claude to navigate, interact, and complete tasks across any browser-based surface, including tools with no API, legacy systems, and sites users are already logged into. The model can work across applications without explicit instruction—reading context from one surface and acting on another, such as checking a calendar, responding to a message, and creating an event without manual orchestration.

For organizations running business workflows on systems predating modern APIs, this browser-based computer use capability is transformative. Developers can also leverage it as a QA and testing layer, delegating visual inspection and form-based validation tasks.

Developer and Enterprise Applications

Software Development Enhancement

For development teams, Claude Sonnet 4.6 delivers a full upgrade to everyday coding workflows. The model is smart enough to work independently through complex codebases and handles iterative workflows without losing quality. Key improvements include:

• Stronger reasoning across code contexts
• Better understanding of complex codebases
• Reliable performance across iterative development cycles

Whether building features, refactoring modules, or debugging issues, Sonnet 4.6 can follow workflows, maintain architectural context, and adapt through iterations. This translates to fewer context resets, faster cycle times, and smoother development velocity.

Knowledge Work and Document Production

Sonnet 4.6 makes high-quality knowledge work accessible at scale, enabling teams to produce polished outputs with fewer editing cycles. The model excels at:

• Drafting and refining reports
• Summarizing large document sets
• Generating structured business documentation
• Producing polished presentations and narratives

Consistent quality across both single-turn tasks and extended multi-turn collaboration ensures teams spend less time refining and more time delivering.

Versatile Use Cases Across Industries

The model supports multiple deployment patterns:

1. Search & Conversational Experiences: Ideal for high-volume conversational products, delivering consistent quality across multi-turn exchanges while remaining cost-efficient at scale.

2. Agentic & Multi-Model Pipelines: Functions as both lead agent and sub-agent in multi-model setups, with adaptive thinking and effort controls providing precise orchestration tools for complex workflows.

3. Finance & Analytics: With stronger financial modeling intelligence and improved spreadsheet capabilities, it's well-suited for analysis, compliance review, and data summarization workflows requiring precision and iteration speed.

4. Enterprise Document Production: Users need fewer rounds of editing to reach production-ready documents, spreadsheets, and presentations, making it valuable for finance, legal, and other precision-critical verticals.

Enterprise-Grade Deployment in Microsoft Foundry

With Claude Sonnet 4.6 available in Microsoft Foundry, customers can deploy near-Opus-level intelligence within an enterprise-grade environment that supports governance, compliance, and operational tooling. This integration provides:

• Enterprise-grade security and compliance frameworks
• Governance controls for AI deployment
• Operational tooling for monitoring and management
• Seamless integration with existing Microsoft infrastructure

For teams building modern AI workflows—from developer assistants to enterprise automation agents—Claude Sonnet 4.6 provides a powerful, scalable foundation within Microsoft's trusted cloud environment.

Getting Started and Learning Resources

The model is available immediately in Microsoft Foundry for enterprise customers. Teams can begin experimenting with the enhanced capabilities right away.

For those seeking deeper understanding, Microsoft is hosting a Model Mondays session on February 23, where leaders from Anthropic will walk through both Claude Opus 4.6 and Claude Sonnet 4.6, including real-world use cases, architectural guidance, and future directions for frontier models in enterprise deployment.

This launch represents Microsoft's continued commitment to providing enterprise customers with access to cutting-edge AI capabilities while maintaining the governance, security, and integration benefits of the Microsoft ecosystem. Claude Sonnet 4.6 fills an important middle tier in the AI model spectrum—delivering performance that approaches the highest tier while remaining accessible for broader deployment across enterprise use cases.