Anthropic has released Claude Opus 4.5, a significant update to its flagship artificial intelligence (AI) model.
Key Improvements
The final member of the Claude 4.5 family boasts major enhancements in code generation, long-context reasoning, and agentic capabilities.
- Code Generation: Opus 4.5 solves more complex coding tasks than its predecessor while using up to 65 percent fewer tokens.
- Long-Context Reasoning: The model can process instructions with less computational cost and within smaller prompt windows, thanks to improved planning and internal reasoning steps.
Capabilities and Advantages
Opus 4.5 is designed to manage complex multi-step workflows, allowing it to refactor codebases, coordinate agent work, and maintain high-level plans while executing low-level details.
- The model's improved long-context reasoning reduces context usage by up to 85 percent.
- Opus 4.5 is capable of producing multi-page narrative chapters (10–15 pages) and maintaining characters, plot direction, and tone across longer passages.
Benchmarks and Performance
Anthropic's internal testing revealed that Opus 4.5 outperformed rivals in code-based tests, including the SWE-Bench Verified benchmark, where it scored 80.9 percent compared to Gemini 3 Pro at 76.2 percent and GPT-5.1 Codex Max at 77.9 percent.
Affordability and Availability
Claude Opus 4.5 is now available across Claude's app, website, API, and major cloud platforms, including Google Vertex AI and Amazon Bedrock, at a cost roughly one-third that of its predecessors for many enterprise workloads.
In conclusion, Anthropic's release of Claude Opus 4.5 marks a significant milestone in the development of artificial intelligence models, offering improved performance and affordability while expanding capabilities in code generation, long-context reasoning, and agentic tool use.