{"version":"2.1.281","anchor":"remote-control-set-max-thinking-tokens-can-now-also-set-the","canonical_anchor":"remote-control-set-max-thinking-tokens-can-now-also-set-the","heading":"Remote Control set_max_thinking_tokens can now also set the thinking display","tier":"notice","area":"Remote Control","url":"https:\/\/changelogs.core-directive.com\/v\/2.1.281\/e\/remote-control-set-max-thinking-tokens-can-now-also-set-the","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.281","markdown":"### Remote Control set_max_thinking_tokens can now also set the thinking display\n\nRemote Control can now set how thinking is displayed along with the thinking budget, unless the local user already chose a display\n\n**What**\n\nRemote Control lets you drive a Claude Code session from another device or app. When it changes the thinking budget (how much the model may reason before answering) through `set_max_thinking_tokens`, it can now also send a thinking display choice. That choice is stored as `thinkingDisplayOverride`.\n\nIf the person at the terminal already picked a display, either with the `--thinking-display` flag or the `showThinkingSummaries` setting, their choice wins. The budget change is still applied, but the display is left as it was, and the remote app receives an error saying the display was not applied.\n\n**Why**\n\nA remote app can now control both how much the model thinks and how that thinking is shown. A display you chose at the terminal is not silently overridden, and the remote side is told when its request was only partly applied.\n\n- Area: Remote Control\n- Names: `set_max_thinking_tokens`\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 2\/5"}