Claude Fable 5.1 requires usage credits on Pro. On Max, it uses your included weekly allowance, subject to a Fable-specific cap. Access to a model and included usage are different things. Anthropic’s plan rules
Is Fable 5.1 included in your plan?
| Plan or seat | How Fable 5.1 is billed |
|---|---|
| Free | Not available |
| Pro | Usage credits; not included usage |
| Max 5x or 20x | Included, up to 50% of the weekly allowance |
| Team standard | Usage credits; not included usage |
| Team premium | Same weekly cap as Max |
| Legacy seat-based Enterprise | Premium: included with the cap; standard: organization-enabled credits |
| Usage-based Enterprise or Claude API | Token billing at API rates |
The 50% cap is part of your existing weekly limit, not an extra allowance. If other models have used 80% of that limit, only 20% remains overall, even if you have barely used Fable. Fable 5 and 5.1 share the Fable allowance. Check the meters in Settings > Usage before starting a long task.
Why can the usage meter move so quickly?
Claude does not promise a fixed number of Fable messages. Conversation length, task complexity, tools, model choice, and effort all affect usage. A short follow-up in a long chat can involve more work than the same question in a fresh conversation. How Claude usage limits work
For routine work, lower the effort setting and leave unnecessary tools off. For an unrelated task, start a fresh conversation rather than carrying a large history forward. Keep the context needed for the answer; removing essential instructions can create more corrections later.
Starting another chat can reduce context overhead, but it does not restore spent usage. A full context window and an exhausted usage allowance are different problems. Check the actual limit notice rather than assuming that every interruption needs a paid upgrade.
What to do when you reach a limit
Open Settings > Usage and check which meter reached its limit and the reset time shown. A short-session reset does not mean your weekly allowance has reset too.
Usage credits are an additional charge calculated at standard API token rates, separate from your subscription fee. To continue with paid usage, check Usage credits on the web. For Pro and Max, you can set a monthly spending cap with Adjust limit, add prepaid funds, and choose whether to enable auto-reload. Review the balance and reload setting separately: permission to spend and available funds are not the same thing. Manage usage credits
Leave usage credits disabled if you do not want additional consumption-based charges. If your plan includes other models and you still have allowance, switch to one of those or wait for the relevant reset. See how subscriptions differ from usage credits.
Why is Fable 5.1 missing or unavailable?
First, confirm the account and workspace you are using, then check its plan against the table above. In Claude on the web, desktop, or mobile, look for Fable 5.1 in the model picker. For Cowork, update Claude Desktop. An organization can restrict model access.
For Claude Code, version 2.1.255 or later is required. Check the installed version, update it if necessary, then reopen the model picker with /model. When connected directly to the Anthropic API, you can also type /model fable: this checks server availability even if the picker has no entry. Confirm that the selected model is Fable 5.1. An old pinned model or an organization restriction can keep a different model selected. Ask your administrator about a restriction instead of changing managed settings. Claude Code model configuration
If Fable appears but will not run, read the error before reinstalling. A billing or usage-limit message needs an account check, not another model-name change. For a wider service problem, check whether the service is down.
Does cheaper API pricing mean more included messages?
No. Anthropic’s cost reduction concerns token-billed work, especially reused context. It is not a promise of a larger subscription quota. Its published Fable 5.1 rates are:
| Token type | USD per million tokens |
|---|---|
| New input | $10 |
| Cached input reads | $0.25 |
| Output | $50 |
These are token rates, not monthly plan prices or a complete bill estimate; cache writes and other applicable charges are separate. The amount of reusable context and output changes the total. Anthropic estimates typical token-billed workloads cost about 25% less than Fable 5, but that is not a guaranteed saving for your task. Official release and pricing
Official sources
- Fable models on your plan: access, seat types, and weekly caps.
- Usage and length limits: shared usage and context limits.
- Usage credit controls: spending limits, funds, and auto-reload.
- Claude Code model configuration: version requirements and model restrictions.
- Fable 5.1 announcement: token pricing and the distinction from Mythos 5.1.
FAQ
Does Fable 5.1 come with a new promotional credit?
No equivalent Fable 5.1 credit is announced. The earlier one-time promotion applied to Fable 5, not this release.
Does switching between Claude Code and the website reset usage?
No. Switching apps does not reset your allowance. Usage on Claude’s website, desktop app, and Claude Code counts toward the same account limits.
Can I enable usage credits through the mobile app?
For Pro and Max, enable and purchase usage credits on Claude’s website, including when your subscription was purchased through an app store.
Does Fable access also give me Mythos 5.1?
No. Anthropic describes Fable 5.1 as generally available, while Mythos 5.1 requires approved access. Do not treat them as interchangeable account entitlements.