Unclear Which models are billed this way is set remotely and is not stated.
What
When you use more than your plan includes, Claude Code may ask your consent to bill extra usage to usage credits. These checks can now be decided separately for each model:
- Consent check: the function that decides whether overage or plan-limit consent is needed now takes an argument. It first looks for an override for that argument and, if one exists, uses that override's
enabledvalue. Otherwise it falls back to the old checks (enabled === false,overageConsentRequired,planLimitsEndDate). - Per-model list: a GrowthBook flag (a remote setting),
tengu_saffron_lattice, now carries aperModelmap of{enabled}entries keyed by model id. Billing status such as needs_consent and needs_credits, and whether the Fable model is disabled in the model picker, are now worked out per model instead of plan-wide. If entries for one model conflict, that model is treated as billed to usage credits.
Nothing has been read about tengu_saffron_lattice yet, which means its state is unknown, not off.
Why
Which models need usage credits can now differ from model to model, so you may see a consent prompt or a disabled entry in the model picker for one model but not another.
tengu_saffron_lattice Not enough to sayNothing here resolved what this flag was doing on this version, so nothing here should be read as on or off.
This account: no value returned · anonymous baseline: no value returned · compiled default in v2.1.299: not a boolean we can read
These values were read against a different version of Claude Code, so treat them as the nearest reading available instead of one taken on this release.
Read once, for one account on one subscription tier, against v2.1.299. It isn't a statement about your account. What a flag value here can and cannot tell you
Which models are billed this way is set remotely and is not stated.