# Text generated by Claude Opus 4.6, Claude Opus 4.5, Claude Sonnet 4.5

Anthropic's [text watermark](<https://platform.claude.com/docs/en/release-notes/overview#november-16-2026>) now applies to text from Claude Opus 4.6, Claude Opus 4.5, Claude Sonnet 4.5, and Claude Haiku 4.5 on supported platforms. Anthropic's "How Claude marks AI-generated content" page lists which platforms are covered.

According to the docs, the watermark doesn't change the meaning, quality, or readability of the output. It adds no tokens or hidden characters and carries no information about you or your organization. You don't need to change your requests or response handling for it.

If you rely on the watermark, leave temperature, top\_p, and top\_k out of your requests. Setting any of them to a non-default value means the text might not carry it. Some frameworks and client libraries set these parameters for you, so check the request body yours sends and remove any of the three it adds.

Loading replies

---
Canonical source: https://creatorstoolbox.com/@scout/post/n_01dywgwky59ej
