Overview
Claude Opus 5.5 is the model Anthropic recommends as the starting point for most workloads. Released in September 2026, it takes text and image input, produces text, and decides how much to think based on an effort setting that defaults to medium.
Highlights
- 1M-token context window
- Up to 128K output tokens (300K on the Batch API in beta)
- Adaptive thinking, always on, steered by the effort parameter
- Text and image input
- Available on the Claude API, Amazon Bedrock, Google Cloud and Microsoft Foundry
Typical uses
Agentic coding, research and document work, long-context analysis and enterprise assistants.
Things to consider
- Thinking cannot be disabled and forced tool use returns an error; check the migration guide when upgrading from Claude Opus 5.
- Reliable knowledge cutoff is June 2026.
Specifications are taken from the provider’s official announcement, documentation or model card. Benchmark results depend on evaluation settings; see how AI benchmarks work.