[DOCS] Model configuration docs still describe Opus 4.6 effort support instead of Opus 4.7 xhigh

Documentation Type

Incorrect/outdated documentation

Documentation Location

https://code.claude.com/docs/en/model-config

Section/Topic

Model aliases / "Adjust effort level"

Current Documentation

The docs currently say:

| sonnet | Uses the latest Sonnet model (currently Sonnet 4.6) for daily coding tasks | | opus | Uses the latest Opus model (currently Opus 4.6) for complex reasoning tasks | Three levels persist across sessions: low, medium, and high. A fourth level, max, provides the deepest reasoning with no constraint on token spending, so responses are slower and cost more than at high. max is available on Opus 4.6 only and does not persist across sessions except through the CLAUDE_CODE_EFFORT_LEVEL environment variable. Effort is supported on Opus 4.6 and Sonnet 4.6. The effort slider appears in /model when a supported model is selected.

The command reference also says:

/effort [low|medium|high|max|auto] ... max applies to the current session only and requires Opus 4.6.

What's Wrong or Missing?

Changelog v2.1.111 says: Claude Opus 4.7 xhigh is now available! Use /effort to tune speed vs. intelligence.

The current Claude Code docs are outdated for this release:

A. The model documentation still identifies Opus 4.6 as the current/latest Opus model

I found no Opus 4.7 or xhigh references in the current code.claude.com mirror, while model-config still says the latest Opus model is Opus 4.6.

B. The effort documentation still scopes the feature to Opus 4.6-era wording

/effort, --effort, settings, and frontmatter docs still describe effort support using "Opus 4.6 only" wording. After v2.1.111, users need documentation that explains how /effort applies to Claude Opus 4.7 xhigh and what "tune speed vs. intelligence" means for that model.

Without that update, users cannot tell whether Opus 4.7 xhigh is a new model picker entry, how it relates to existing effort levels, or which effort options are supported on the new model.

Suggested Improvement

Update the model and effort documentation for v2.1.111:

1. In model-config, replace stale "currently Opus 4.6" / "Opus 4.6 only" wording with the current Opus 4.7 xhigh behavior.
2. Add a short explanation of where users will see Claude Opus 4.7 xhigh (for example, in /model) and how /effort changes the speed vs. intelligence tradeoff on that model.
3. Clarify whether xhigh is a model variant name, an effort preset, or a label shown alongside existing effort levels so users do not confuse it with high / max.
4. Update the command, settings, env-var, skill, and subagent references that still describe effort support as Opus 4.6-specific if that restriction is no longer accurate.

Impact

Medium - Makes feature difficult to understand

Additional Context

Affected Pages:

| Page | Line(s) | Context |
|------|---------|---------|
| https://code.claude.com/docs/en/model-config | 39-40, 157-176 | Alias table still says latest Opus is 4.6; effort support and max are described with Opus 4.6-specific wording |
| https://code.claude.com/docs/en/commands | 40 | /effort says max requires Opus 4.6 |
| https://code.claude.com/docs/en/cli-reference | 67 | --effort says max is Opus 4.6 only |
| https://code.claude.com/docs/en/common-workflows | 525-541 | Effort tuning is documented only for Opus 4.6 and Sonnet 4.6 |
| https://code.claude.com/docs/en/settings | 186 | effortLevel says it is supported on Opus 4.6 and Sonnet 4.6 |
| https://code.claude.com/docs/en/env-vars | 87 | CLAUDE_CODE_EFFORT_LEVEL says max is Opus 4.6 only |
| https://code.claude.com/docs/en/skills | 199 | Skill frontmatter still says max is Opus 4.6 only |
| https://code.claude.com/docs/en/slash-commands | 199 | Slash-command frontmatter still says max is Opus 4.6 only |
| https://code.claude.com/docs/en/sub-agents | 255 | Subagent frontmatter still says max is Opus 4.6 only |

Total scope: 9 pages affected

Source: Changelog v2.1.111

Exact changelog entry: Claude Opus 4.7 xhigh is now available! Use /effort to tune speed vs. intelligence