[DOCS] Model configuration page says Opus 4.6 and Sonnet 4.6 default to medium effort for all providers

Documentation Type

Incorrect/outdated documentation

Documentation Location

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

Section/Topic

Adjust effort level

Current Documentation

The page currently says:

Effort levels control adaptive reasoning, which dynamically allocates thinking based on task complexity. Lower effort is faster and cheaper for straightforward tasks, while higher effort provides deeper reasoning for complex problems. Opus 4.6 and Sonnet 4.6 default to medium effort. This applies to all providers, including Bedrock, Vertex AI, and direct API access. Medium is the recommended level for most coding tasks: it balances speed and reasoning depth, and higher levels can cause the model to overthink routine work.

What's Wrong or Missing?

This is outdated as of changelog v2.1.94.

The release note says:

Changed default effort level from medium to high for API-key, Bedrock/Vertex/Foundry, Team, and Enterprise users (control this with /effort)

That contradicts the current documentation's blanket statement that Opus 4.6 and Sonnet 4.6 default to medium effort for all providers.

Because /effort auto resets to the model default, this is not just wording drift. It changes what users should expect for reasoning depth, latency, and cost depending on how they access Claude Code.

Suggested Improvement

Update the Adjust effort level section to describe the current default by access mode or plan instead of claiming a universal medium default.

At minimum, document that v2.1.94 changed the default to high for:

• API-key users
• Bedrock, Vertex AI, and Microsoft Foundry users
• Team and Enterprise users

Also clarify that /effort can override the default and that /effort auto restores the applicable default for the current access mode or plan.

If any remaining cohorts still default to medium, list them explicitly rather than saying the same default applies to all providers.

Impact

Medium - Makes feature difficult to understand

Additional Context

Affected Pages:

| Page | Context |
|------|---------|
| https://code.claude.com/docs/en/model-config | Primary effort-default explanation; currently says medium is the default for all providers |
| https://code.claude.com/docs/en/commands | /effort ... auto says it resets to the model default, so users need the default documented correctly |
| https://code.claude.com/docs/en/env-vars | CLAUDE_CODE_EFFORT_LEVEL=auto also depends on the documented model default |
| https://code.claude.com/docs/en/common-workflows | Extended thinking guidance sends readers to the effort-level documentation |

Total scope: 4 pages affected

Source: Changelog v2.1.94

Exact changelog entry:

Changed default effort level from medium to high for API-key, Bedrock/Vertex/Foundry, Team, and Enterprise users (control this with /effort)