Model configuration changedmodel-config
Upstream edited this page at 28 Oct 2026 19:44 UTC, give or take a minute or two: the time comes from Anthropic’s own sitemap rather than from a commit. This site recorded the change at 28 Oct 2026 20:07 UTC.
Upstream edited
Recorded here
Lines+18added
Lines−1removed
From line
685
where the diff opens
First seen
14 Aug 2026
this site's first read of the page
Recorded edits84to this page, all time
#### Send no effort level
The whole hunk
from line 685, old and new numbered
/
from line 685
685685* **`/effort`**: run `/effort` with no arguments to open an interactive slider, `/effort` followed by a level name to set it directly, or `/effort auto` to clear your saved level for the active model
686686* **In `/model`**: use left/right arrow keys to adjust the effort slider when selecting a model
687687* **`--effort` flag**: pass a level name to set it for a single session when launching Claude Code
688* **Environment variable**: set `CLAUDE_CODE_EFFORT_LEVEL` to a level name or `auto`
688* **Environment variable**: set `CLAUDE_CODE_EFFORT_LEVEL` to a level name. To have Claude Code [send no effort level](#send-no-effort-level), set it to `auto` or `unset` instead
689689* **Settings**: set a per-model level in [`modelSettings`](https://code.claude.com/docs/en/settings-reference#modelsettings), or set [`effortLevel`](https://code.claude.com/docs/en/settings-reference#effortlevel) to `low`, `medium`, `high`, or `xhigh` as the default for models without one. `max` isn't accepted as a level in either key, and `ultracode` has its own [`ultracode`](https://code.claude.com/docs/en/settings-reference#ultracode) key
690690* **From a connected device**: in a [Remote Control](https://code.claude.com/docs/en/remote-control.md#what-connected-devices-see) session, pick a level from the effort control on your phone or in your browser. The level applies to the current session only. Requires Claude Code v2.1.234 or later
691691* **Skill and subagent frontmatter**: set `effort` in a [skill](https://code.claude.com/docs/en/skills.md#frontmatter-reference) or [subagent](https://code.claude.com/docs/en/sub-agents.md#supported-frontmatter-fields) markdown file to override the effort level when that skill or subagent runs
from line 706
706706* **While Claude is still working**: type a message, such as `keep going`, and press `Enter` to queue it without interrupting the turn. Claude gets it when its current tool calls finish, or when the turn ends at the latest, and works at the new level from then on.
707707
708708Before v2.1.242, Claude Code could queue `/effort` itself until the turn finished, and always did in a session that doesn't [fetch feature flags](https://code.claude.com/docs/en/env-vars.md#features-that-need-feature-flag-fetching), such as on a [third-party provider](https://code.claude.com/docs/en/third-party-integrations.md).
709
710#### Send no effort level
711
712When Claude Code sends no effort level to a [supported model](#adjust-effort-level), the [API's default](https://platform.claude.com/docs/en/build-with-claude/effort) applies instead of Claude Code's. To have Claude Code send none, start `claude` from a shell that exports `CLAUDE_CODE_EFFORT_LEVEL` with the value `auto` or the value `unset`, which do the same thing:
713
714```bash theme={null}
715export CLAUDE_CODE_EFFORT_LEVEL=auto
716```
717
718The session header and footer show no effort level while none is sent. To confirm that none is sent, run `/effort status` at the Claude Code prompt:
719
720* **`Effort level: auto`**: no effort level is sent.
721* **`Effort level: auto (currently <level>)`**: a level is still sent. If a [`maxEffortLevel`](https://code.claude.com/docs/en/settings-reference#maxeffortlevel) setting below `max` applies to the active model, Claude Code sends its own default for the model, lowered to that cap. Remove the setting or set it to `max`.
722
723> [!NOTE]
724>
725> Before v2.1.304, `/effort status` printed `Effort level: auto (currently high)` even while no level was sent.
709726
710727#### Adaptive reasoning and fixed thinking budgets
711728
No line in this hunk matches that.