Group of 2 You'll notice
Claude Code now re-probes the server before blocking a model it previously denied access to, letting access restore itself without a restart.
What
- Claude Code now tracks models the server previously denied access to (
confirmedEntitlementDenies). - Before re-reporting a denial or finalizing a block, it re-probes the server (with a deadline, forcing a fresh server check) to see whether the deny has since been overturned — for example because your organization or account changed.
- If the deny is overturned, the cache is updated and the model is evicted from the deny list, so startup or the
/modelcommand can resolve to the now-available model without restarting Claude Code. - A blocked result now also carries an
underEnforcementflag, and failures during this re-probe are tracked distinctly asentitlement_probe_failedversus a normalnot_alloweddenial.
Why Previously a denied model could stay blocked even after the underlying restriction was lifted, requiring a restart. Re-probing lets access self-heal automatically.
Names in the bundle/model
/model
Claude apps gateway for Amazon Bedrock, Claude Platform on AWS, Google Cloud, and Microsoft Foundry modified, high confidence
Once the developer signs in, the [model picker](/docs/en/model-config) shows the models in their `availableModels` allowlist. Managed settings apply at startup and refresh hourly, and telemetry routes to your collector.see the edit
/model
All settings modified, high confidence
See [Ask before switching](/docs/en/model-config#ask-before-switching).see the edit
/model
Environment variables modified, high confidence
| `CLAUDE_CODE_USER_DIALOG_TIMEOUT_MS` | Deadline in milliseconds before Claude Code cancels a dialog it forwards to a remote client such as a [Remote Control](/docs/en/remote-control) or SDK host, or the approval dialog for a [held cross-…see the edit
/model
All settings modified, high confidence
Set the deadline for dialogs Claude Code [forwards to a remote client](/docs/en/remote-control#limitations), such as a Remote Control or SDK host, and for the approval dialog for a [held cross-session message](/docs/en/cross-session-messag…see the edit
/model
Claude Code changelog modified, high confidence
* Fixed switching models with /model re-sending every tool definition (a prompt-cache miss); commit and PR attribution text now arrives as a conversation note that updates on model changessee the edit
/model
Error reference modified, high confidence
You picked a model with `/model <name>` and Claude Code couldn't confirm that a model with that name exists. When the name isn't a [model alias](/docs/en/model-config#model-aliases) or another spelling Claude Code accepts locally, `/model`…see the edit
/model
Set up Claude Code for your organization modified, high confidence
| [Effort cap](/docs/en/settings-reference#maxeffortlevel) | Cap the [effort level](/docs/en/model-config#adjust-effort-level) for every model or per model, on every provider | `maxEffortLevel` |see the edit
/model
Escalate hard decisions with the advisor tool modified, high confidence
* **Subscription plans**: advisor usage counts toward your plan's usage limits, except that a Fable advisor bills to [usage credits](/docs/en/model-config#fable-and-usage-credits) on plans where Fable usage doessee the edit
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.
Added since
A small documentation edit on Deploy Claude Desktop on 3P with an LLM gateway touched a line naming /model after this was published.
If your gateway serves a Claude model under an opaque routing alias, it can mark the model as Claude by returning an `anthropic_family_tier` field (a Claude tier name such as `sonnet` or `opus`) on that model object in its `/v1/models` res…third-party/claude-desktop/gateway see the edit
Confirmed since
Anthropic's documentation has since written up /model, on Claude Code changelog.
* Added `CLAUDE_CODE_GATEWAY_MODEL_DISCOVERY_TIMEOUT_MS` to extend the LLM gateway `/v1/models` discovery timeout (default 3s)changelog see the edit
One source agreesOne thing we can check says the same as this entry.
Anthropic's documentation agrees
/model on Claude apps gateway for Amazon Bedrock, Claude Platform on AWS, Google Cloud, and Microsoft Foundry