[DOCS] Settings docs missing `modelOverrides` configuration for custom provider model IDs

Resolved 💬 3 comments Opened Mar 11, 2026 by coygeek Closed Mar 16, 2026

Documentation Type

Missing documentation (feature not documented)

Documentation Location

https://code.claude.com/docs/en/settings

Section/Topic

settings.json model configuration keys for mapping model-picker entries to provider-specific model IDs.

Current Documentation

The settings reference currently documents model controls as:

| model | Override the default model to use for Claude Code | | availableModels | Restrict which models users can select via /model, --model, Config tool, or ANTHROPIC_MODEL. |

No documentation currently exists for a modelOverrides setting.

What's Wrong or Missing?

Changelog v2.1.73 introduced modelOverrides to map model picker entries to custom provider model IDs (for example, Bedrock inference profile ARNs).

Without a documented schema and examples, teams using Bedrock, Vertex, or Foundry cannot reliably configure this new setting or understand how it interacts with existing model controls.

Suggested Improvement

Add a dedicated modelOverrides section to Settings (and cross-link from Model configuration) covering:

  • Key shape/type and valid override targets
  • Provider-specific examples (for example Bedrock inference profile ARN)
  • Precedence/interaction with model, availableModels, and ANTHROPIC_DEFAULT_* environment variables
  • How overrides affect /model picker behavior

Impact

Medium - Makes feature difficult to understand

Additional Context

Affected Pages:

| Page | Context |
|------|---------|
| https://code.claude.com/docs/en/settings | Primary settings.json key reference where modelOverrides should be documented |
| https://code.claude.com/docs/en/model-config | Model alias/name mapping behavior should cross-reference settings-based overrides |
| https://code.claude.com/docs/en/amazon-bedrock | Bedrock users need inference-profile mapping examples for custom provider IDs |
| https://code.claude.com/docs/en/third-party-integrations | Enterprise deployment guidance should mention this settings-based mapping option |

Total scope: 4 pages affected

Source: Changelog v2.1.73

Exact changelog entry:

Added modelOverrides setting to map model picker entries to custom provider model IDs (e.g. Bedrock inference profile ARNs)

View original on GitHub ↗

This issue has 3 comments on GitHub. Read the full discussion on GitHub ↗