Follow Discord
Sweep 22 Sep 2026 · 17:19Z Build v2.1.280 501 read Stable v2.1.267 Latest v2.1.280 Next v2.1.280 Feeds RSS JSON llms.txt Unofficial
Claude Code v2.1.267 ·

New maxEffortLevel setting caps reasoning effort

Administrators can now cap the maximum reasoning-effort level, overall or per model, and Claude Code enforces it against any effort request

TierUse it nowhow much it should matter to you
Useful4my rating, 1 to 5
Signal3worth watching, 1 to 5
AreaEffortwhat it touches
KindNew Featuresin v2.1.267,
Group of 3 Use it now Notable

Administrators can now cap the maximum reasoning-effort level, overall or per model, and Claude Code enforces it against any effort request

What

  • A new managed-settings key, maxEffortLevel, lets an administrator restrict the maximum reasoning-effort level allowed.
  • The effective cap is computed by combining a top-level maxEffortLevel with any per-model modelSettings.<model>.maxEffortLevel overrides across all loaded settings files, taking the lowest applicable value.
  • Any explicit effort request, from /effort, /model, the --effort flag, the CLAUDE_CODE_EFFORT_LEVEL environment variable, or a model's default, is clamped to this cap.
  • Effort-level resolution (whether effort, max_effort, or xhigh_effort apply for a given model) now checks this new setting first, before falling back to the older logic based on the model's id.

Why This lets organizations enforce a ceiling on how much reasoning effort, and therefore cost and latency, Claude Code can use, regardless of what an individual user or model requests.

Read from
Names in the bundlemaxEffortLevel
Since it was published

The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.

Confirmed since Anthropic's documentation has since written up maxEffortLevel, on What's new. Also this week: pop any **Claude Code Desktop pane** out into its own window and dock it back later; the **`maxEffortLevel`** setting caps the effort level on every provider; and a page that **WebFetch** hasn't finished downloading within … whats-new/index see the edit
Added since A small documentation edit on Manage multiple agents with agent view touched a line naming /effort after this was published. If the session took its effort from your settings rather than from `--effort` or `/effort`, Claude Code reads your settings again each time it starts a process for the session. After you edit the saved effort in `settings.json`, the change… agent-view see the edit
Added since A small documentation edit on Manage multiple agents with agent view touched a line naming --effort after this was published. If the session took its effort from your settings rather than from `--effort` or `/effort`, Claude Code reads your settings again each time it starts a process for the session. After you edit the saved effort in `settings.json`, the change… agent-view see the edit
Confirmed since Anthropic's documentation has since written up --effort, on CLI reference. | `claude agents` | Open [agent view](/docs/en/agent-view) to monitor and dispatch parallel background sessions. Use `--cwd <path>` to show only sessions started under that directory, or `--json` to print active sessions as a JSON array fo… cli-reference see the edit
How sure we are
Two sources agreeTwo things we can check say the same as this entry.
Anthropic's documentation agreesAnthropic's documentation has since written up --effort, on CLI reference.
Anthropic's release notes agreeAdded maxEffortLevel setting (top-level or per model under modelSettings): caps the effort level on every provider, including Bedrock…
The name it cites is new in this buildNew in this build: maxEffortLevel

See this entry in the whole of v2.1.267 →

Feedback