claude-opus-5-5-1009
Claude Opus 5.5 is the first high-capability model in Anthropic’s Claude 5.5 family. It is designed for complex software engineering, long-horizon agents, professional research, and enterprise knowledge work. It excels at cross-file code migrations, code review, bug fixing, data analysis, financial modeling, report writing, and multi-step automation, while maintaining planning, tool use, and verification throughout extended workflows. Compared with Claude Opus 5, Opus 5.5 improves task completion, execution efficiency, and output quality on complex work, making it particularly well suited to coding and business processes that require long periods of autonomous operation. Anthropic has also strengthened its safety mechanisms, including protections against prompt injection, unauthorized actions, and high-risk tasks. The model uses adaptive thinking to allocate reasoning resources according to task difficulty, while offering lower typical workload costs and faster generation than Opus 5. It is well suited to enterprise use cases that demand reliability, deep reasoning, and high-quality deliverables.
- Context
- 1000000 tokens
- Max output
- 128000 tokens
- Input
- llm image
- Output
- llm
Pricing
Input: $4 / 1M tokens
Output: $20 / 1M tokens
Cached input: $0.2 / 1M tokens
Cache write: $5 / 1M tokens
Endpoints
- anthropic
- openai