Claude Models API adds a check for turning thinking off

Developers can inspect support before selecting a thinking configuration.

Jordan Reid
October 6, 2026
Tools, Anthropic
1 min

The Claude website displayed on a screen — An illustrative photograph of the Claude website. Photo by Planet Volumes / Unsplash.

Listen to this article

Anthropic added a capability check to the Claude Models API on October 5. Applications can now ask whether a model accepts requests that turn thinking off.

The new field is capabilities.thinking.types.disabled. It appears in responses from both the model list and individual model endpoints.

The documentation says a false supported value means the model rejects the disabled setting with a 400 error. Models without thinking support report true.

A true value still leaves other constraints to check. A request can fail when its selected effort level is incompatible with thinking being off.

For developers maintaining model selectors, the field provides a direct compatibility signal. Applications still need to validate the complete request configuration.

Archival photograph of the Claude website by Planet Volumes on Unsplash, published April 6, 2026 and used under the Unsplash License. The image illustrates Claude and does not show the October Models API update. Existing WebP rendition.

Jordan Reid

Jordan Reid is focused on AI tools, agents, developer products, and the way technology changes everyday work. Jordan approaches a launch from the user’s side of the screen. What can it actually help someone finish? The voice is practical, conversational, and skeptical of products that turn a simple job into five new settings. Coverage follows coding assistants, creative software, browser agents, and the workflows around them, with attention to pricing, permissions, setup, and the human work that remains.