Model Settings

These keys choose the model Claude Code uses and how hard it thinks. Administrators also use them to allow or block models. Several keys have a matching environment variable that wins over the key.

Use /model or edit a settings file. fastMode, alwaysThinkingEnabled, and enforceAvailableModels switch features.

Keys

Source: Settings Reference for rows marked

Documented: An official docs page names this item.

. For rows marked

Undocumented: Only the program shows this item. It can change without notice.

, Source: the program (its bundled settings schema). Where to Set It tells you which files may set the key. See Settings Files and Scopes.

Key	Status	Type	Default	What It Does	Where to Set It
`advisorModel`	Documented: An official docs page names this item.	string	-	Pick which model answers when Claude asks the advisor tool.	Any settings file
`alwaysThinkingEnabled`	Documented: An official docs page names this item.	boolean	-	Turn extended thinking off for every session.	Any settings file
`availableModels`	Documented: An official docs page names this item.	array of string	-	Restrict which models people can pick. It accepts family aliases (`opus`), version prefixes (`opus-4-5`), and full model IDs. An empty array leaves only the default model.	Any settings file
`availableModelsMatch`	Documented: An official docs page names this item. Managed Only: Works only in managed (organization policy) settings.	enum: `prefix`, `exact`	-	Make each `availableModels` model ID entry permit only the version it names. `prefix` is the default. `exact` stops an entry from allowing other versions.	Managed settings only
`deniedModels`	Documented: An official docs page names this item. Managed Only: Works only in managed (organization policy) settings.	array of string	-	Block specific models, even ones `availableModels` permits. A family alias blocks the family. Claude Code does not start if no allowed model is left for Default.	Managed settings only
`effortLevel`	Documented: An official docs page names this item.	enum: `low`, `medium`, `high`, `xhigh`	-	Set a default effort level for models without a saved level of their own. Claude Code saves levels for supported models.	Any settings file
`enforceAvailableModels`	Documented: An official docs page names this item.	boolean	-	Keep the `/model` Default choice inside your `availableModels` allowlist. It has no effect when `availableModels` is unset or empty.	Any settings file
`fallbackModel`	Documented: An official docs page names this item.	array of string	-	Name backup models for when the primary is overloaded. Claude Code tries the models in order. The `--fallback-model` flag wins. `default` expands to the default model.	Any settings file
`fastMode`	Documented: An official docs page names this item.	boolean	-	Turn fast mode on for sessions where it's available.	Any settings file
`fastModePerSessionOptIn`	Documented: An official docs page names this item.	boolean	-	Require people to turn fast mode on each session. Each session starts with fast mode off.	Any settings file
`maxEffortLevel`	Documented: An official docs page names this item.	string	-	Cap the effort level for every model or per model, on every provider. The lowest value across files wins. Claude Code does not clamp an effort set through `CLAUDE_CODE_EXTRA_BODY`.	Any settings file
`model`	Documented: An official docs page names this item.	string	-	Change the model Claude Code starts with.	Any settings file
`modelOverrides`	Documented: An official docs page names this item.	object	-	Map model IDs to your provider's IDs, such as Bedrock ARNs. Usually set in managed settings. Since `2.1.305`, `/effort` and `/autocompact` settings apply even when the model name differs from a `modelOverrides` value by letter case or a `[1m]` tag (v2.1.305 release).	Any settings file
`modelPicker`	Documented: An official docs page names this item.	object	-	Choose which models the `/model` picker lists, in your own order and with your own labels. Claude Code reads it only from managed, `--settings`, and user settings. The first source that sets it wins whole.	User or managed settings
`modelPricing`	Documented: An official docs page names this item. Managed Only: Works only in managed (organization policy) settings.	object	-	Report spend at your organization's contracted rates instead of list price. `overrides` maps a model ID to USD rates per million tokens. The figures stay estimates.	Managed settings only
`modelSettings`	Documented: An official docs page names this item.	object: model name to `{effortLevel, maxEffortLevel, autoCompactWindow}`	-	Keep a saved effort level (`low`, `medium`, `high`, `xhigh`), a top effort cap (`maxEffortLevel`), or an auto-compact window (`"auto"` or 100000 to 1000000 tokens) for each model. `/autocompact` saves here. Claude Code drops invalid values. Across settings files the lowest cap wins. Source: the program.	Any settings file
`promptCacheTtl`	Documented: An official docs page names this item.	string	-	Choose the prompt cache lifetime for the main conversation. The values are `5m` or `1h`. When unset, it is 1 hour on a subscription and 5 minutes on an API key or cloud provider. `CLAUDE_CODE_PROMPT_CACHE_TTL` wins.	Any settings file
`showThinkingSummaries`	Documented: An official docs page names this item.	boolean	-	See summaries of Claude's thinking instead of a collapsed stub.	Any settings file
`subagentPromptCacheTtl`	Documented: An official docs page names this item.	string	-	Choose the prompt cache lifetime for subagent requests. The values are `5m` or `1h`. When unset, it is 5 minutes unless `ENABLE_PROMPT_CACHING_1H=1`. `CLAUDE_CODE_SUBAGENT_PROMPT_CACHE_TTL` wins.	Any settings file
`switchModelsOnFlag`	Documented: An official docs page names this item.	boolean	-	Switch models automatically or pause when a safety classifier flags a request.	Any settings file
`totalTokensReminder`	Undocumented: Only the program shows this item. It can change without notice.	enum: `off`, `infinite`, `fixed`, `countdown`, `padded-countdown`	`padded-countdown`	Add a "tokens left" reminder to the system prompt and after each tool result. Modes: `off`, `infinite`, `fixed`, `countdown`, `padded-countdown`. `CLAUDE_CODE_TOTAL_TOKENS_REMINDER` wins.	Settings files
`totalTokensReminderAfterUserTurn`	Undocumented: Only the program shows this item. It can change without notice.	boolean	On	Repeat the "tokens left" reminder after each user prompt. When off, it shows only in the system prompt and after tool results. `CLAUDE_CODE_TOTAL_TOKENS_REMINDER_AFTER_USER_TURN` wins.	Settings files
`totalTokensReminderBudget`	Undocumented: Only the program shows this item. It can change without notice.	integer	`15000000`	Set the starting token budget of `padded-countdown`. `CLAUDE_CODE_TOTAL_TOKENS_REMINDER_BUDGET` wins.	Settings files

Example

Opus at High Effort with a Fallback

This file starts each session on Opus at high effort. If Opus is overloaded, Claude Code tries Sonnet. The main conversation keeps its prompt cache for one hour.

~/.claude/settings.json
{
  "$schema": "https://clawdipedia.cc/schemas/claude-code-settings.json",
  "model": "opus",
  "effortLevel": "high",
  "fallbackModel": ["sonnet"],
  "promptCacheTtl": "1h"
}

Settings Files and Scopes explains the scopes, precedence, and managed settings.
Settings Reference is the official page for each key.
Model and Effort Variables lists the variables that win over these keys.
Models explains how Claude Code makes model, effort, and retry choices.
Models and Providers lists the model catalog and aliases.
Model Configuration explains aliases and effort in the official docs.

Keys​

Example​

Opus at High Effort with a Fallback​

Related Pages​

Keys

Example

Opus at High Effort with a Fallback

Related Pages