How Anthropic’s Claude 4.5 Targets Enterprise Automation

Software development is fast becoming a primary competitive arena for AI firms, as models gain powerful capabilities to write code and operate computer systems autonomously. For the telecommunications sector, where complex software supports everything from network management to customer billing, these advancements present a significant opportunity for transformation.

Anthropic has released Claude Sonnet 4.5, a new frontier model engineered explicitly for sophisticated software development and computer operation tasks. The launch arrives alongside a suite of product updates, including checkpoints in the Claude Code command-line tool and expanded capabilities within its consumer applications.

Crucially for enterprise development teams, the release includes the Claude Agent SDK, which provides the foundational infrastructure for building bespoke, agentic AI systems.

A toolkit for advanced agentic systems

Anthropic designed Claude Sonnet 4.5 to serve as the core of intelligent automation. The accompanying Agent SDK gives developers a robust framework for creating powerful agent systems.

The toolkit provides essential systems for managing memory across multiple tasks, handling granular permission controls and coordinating the actions of several agents working in concert. Anthropic now makes the same infrastructure it uses to build Claude Code available to organisations building their own solutions.

The strategic importance of mastering code-based environments is evident. “Code is everywhere. It runs every application, spreadsheet and software tool you use,” Anthropic explains. “Being able to use those tools and reason through hard problems is how modern work gets done.”

For telecommunications providers, these capabilities open new avenues for automating network configuration, optimising operational support systems (OSS) and accelerating the development of new services.

The model is available through the Claude API using the identifier claude-sonnet-4-5. It maintains the exact competitive pricing as its predecessor, Claude Sonnet 4, at US$3 per million input tokens and US$15 per million output tokens, ensuring accessibility for large-scale deployments.

Delivering measurable performance gains

Claude Sonnet 4.5 exhibits a substantial performance improvement on industry-standard benchmarks. The model achieves a score of 61.4% on OSWorld, a comprehensive test that evaluates an AI’s ability to perform complex computer tasks.

The score represents a significant increase from the 42.2% achieved by Claude Sonnet 4 just four months prior. Furthermore, on the SWE-bench Verified evaluation, which measures software coding and debugging abilities, Claude Sonnet 4.5 currently leads among all tested models.

Beyond benchmarks, the model exhibits remarkable endurance for complex, multi-step processes. Anthropic reports that the model can maintain focus for more than thirty hours on a single task, a critical feature for managing prolonged network diagnostics or large-scale software deployments.

Industry endorsements highlight real-world impact

Early customers are already reporting impressive results from integrating the new model into their workflows. Mario Rodriguez, Chief Product Officer (CPO) at GitHub, notes the model’s synergy with their established tools.

“Claude Sonnet 4.5 amplifies GitHub Copilot’s core strengths,” he says. “Our initial evals show significant comprehension – enabling Copilot’s agentic experiences to handle complex, codebased-spanning tasks better.”

The benefits extend to enhancing developer productivity within large, intricate software environments.

Eric Wendelin, Tech Lead for Gen AI for Developer Productivity at Netflix, adds: “Claude Sonnet 4.5 is excellent at software development tasks, learning our codebase patterns to deliver precise implementations. It handles everything from debugging to architecture with deep contextual understanding, transforming our development velocity.”

For telecommunications organisations, such endorsements highlight the potential for accelerating development cycles for internal OSS/BSS platforms and network management tools.