Which models Claude Max gives you
Which Claude models each paid plan includes, how Fable's cap fits inside your shared weekly allowance, and what happens when you reach a limit.
Which models Claude Max gives you
Claude Max gives you the same model families that Claude Pro reaches. The higher tier buys headroom rather than a different product: more work before you run out, and a much larger share of that work on the most capable model. Model access is set by your plan, and Fable has an additional cap within the shared weekly allowance; either limit can run out first.
Two rules explain today's access, so learn them before you learn any model name. Your plan decides which models appear in the picker and how much of each you get. Fable carries an additional allowance cap, but other models draw from the same weekly allowance, so reaching the Fable cap first is not guaranteed.
The model names below are dated on purpose. They change every few months. The mechanics can change too.
The model list, checked on 10 November 2026
On Anthropic's plan and pricing page, checked on 10 November 2026, the Pro, Max and Ultra rows list the same four model families: Fable, Opus, Sonnet and Haiku. That page shows Max at 100 USD per month and Ultra at 200 USD per month, with monthly billing only for both plans. Anthropic's Max plan help article, checked the same day, gives Max five times Pro's usage every five hours and every week, and notes that mobile pricing may vary by app platform. Anthropic's Ultra plan help article, also checked that day, says Ultra, formerly Max 20x, gives ten times Pro's usage every week, that existing subscribers move automatically with no price change, and that the 200 USD price is for web subscriptions. Max and Ultra are separate plans, and both cost 20 USD per Pro-sized weekly allowance. The gap between them is allowance, not catalogue.
On Anthropic's model overview, checked on 10 November 2026, the current lineup is Claude Fable 5.2, Claude Opus 5.5, Claude Sonnet 5.5 and Claude Haiku 5.5, and the page tells you to start with Claude Opus 5.5 for most workloads. Fable 5.2 launched on 9 November 2026 and replaced Fable 5.1 in that lineup. The same page lists Claude Fable 5.1, Claude Fable 5, Claude Opus 5, Claude Sonnet 5 and Claude Haiku 4.5 among the legacy models that are still available, alongside older builds including Claude Opus 4.8 and Claude Sonnet 4.6. In Claude Code v2.1.321 and later, the fable alias resolves to Fable 5.2.
Read both pages yourself before you pay for anything. This guide will be old at some point, and the model at the top of the list today becomes a legacy entry later. Anthropic's own rows are the only source that is current by definition.
A couple of access details are conditions rather than names, and they catch people out:
- Opus 5.5 and Opus 5 run with a 1 million token context window on every paid plan, Pro included, according to Anthropic's context window article, checked on 10 November 2026. The one Opus option that still needs usage credits on Pro is the legacy Opus 4.6 1M variant in Claude Code,
claude-opus-4-6[1m]. - Fable requests can bill to usage credits instead of your plan's included limits, depending on your plan and seat tier. On Pro, Fable normally runs on usage credits from the start, while Max and Ultra let you spend up to 50% of your weekly usage limits on Fable at no extra cost before usage credits take over. For the Fable 5.2 launch week, Pro includes Fable up to 50% of its weekly limits from 9 November until 16 November 2026 at 11:59 pm PT, after which Fable goes back to usage credits. In an interactive Claude Code session, a consent prompt appears before Fable bills usage credits; once you choose to continue on Fable with usage credits, it does not appear again.
If you are still choosing between Max and Ultra, the price and allowance comparison is a post of its own: Max and Ultra compared side by side. What a single subscription stretches across is covered in what one Claude Max plan actually includes.
Why the top model runs out first
Max is metered in a five-hour session window and a weekly window. Verified Ultra accounts have no five-hour limit; accounts not yet verified have a temporary limit. Anthropic's usage guidance, checked on 10 November 2026, tells you to check when your plan's weekly usage limit resets "for all models, and for Fable (if included in your plan)", and that phrase is the whole mechanic: one weekly limit covers every model, and Fable carries a cap of its own inside it. On Max and Ultra, that cap is "up to 50% of your weekly usage limits on Fable models at no extra cost". Every other model draws on the same weekly limit, so the Fable share does not add to it, and once the Fable cap is spent you either switch model or continue on Fable with usage credits. Switching only helps while shared allowance remains; the shared weekly limit can run out before the Fable cap. Claude Code also still prints model-family messages for Opus and Sonnet, covered in the next section.
The published list prices behind these models show the cost difference.
The data behind this chart
[
{
"label": "Haiku 5.5",
"input_usd_per_mtok": 0.1,
"output_usd_per_mtok": 0.5
},
{
"label": "Sonnet 5.5",
"input_usd_per_mtok": 2,
"output_usd_per_mtok": 10
},
{
"label": "Opus 5.5",
"input_usd_per_mtok": 4,
"output_usd_per_mtok": 20
},
{
"label": "Fable 5.2",
"input_usd_per_mtok": 8,
"output_usd_per_mtok": 40
}
]Those 4 models are the current lineup at Anthropic's published API list prices, checked on 10 November 2026. Opus 5.5 output lists at 20 USD per million tokens, against 10 USD for Sonnet 5.5 and 0.5 USD for Haiku 5.5 on prompts up to 100,000 tokens, with longer Haiku 5.5 prompts paying 2.50 USD. Fable 5.2 is the most expensive model in the current lineup at 40 USD, and what Fable is actually worth using for explains when that price earns itself. The legacy models keep their own prices: Fable 5.1 and Fable 5 output is 50 USD, Opus 5 is still 25 USD, Sonnet 5 is 10 USD and Haiku 4.5 is 5 USD. Fable uses the shared weekly allowance faster than other Claude models. This is also why matching the model to the task beats leaving the strongest one selected by default: which model to reach for on which job goes through it case by case.
What Claude Code prints when the allowance is gone
Claude Code's subscription allowance messages include these four, and they do not mean the same thing.
You've hit your session limitandYou've hit your weekly limitare plan-wide. They count across every model, so switching model with/modeldoes not restore access.You've hit your Opus limitandYou've hit your Sonnet limitapply only to that model family. Switching to a model outside that family with/modelkeeps you working.
Other messages cover spending limits, including You've hit your monthly spend limit, individual and organization limits, shared budgets, and gateway spending caps. These require checking the named budget rather than switching models.
That distinction answers the question most people are really asking. Claude Code does not quietly move you to a lighter model when your Opus allowance is gone. It stops, prints the message, and tells you when the window resets. You then pick one of these:
- Run
/modeland choose a model outside the exhausted family. Sonnet or Haiku still work while the Opus counter is empty. - Wait for the reset. On Claude Code v2.1.234 and later the session can wait and then continue the interrupted task on its own; when it does not start that wait itself, you pick it from
/rate-limit-options. - Turn on usage credits with
/usage-creditsand keep going past the plan allowance at a per-token price. - If your account has an eligible limit reset, use it from Settings > Usage in Claude on the web, Claude Desktop or the Claude mobile app; the reset button is not available in Claude Code in the terminal or IDE. A five-hour or weekly reset refills that limit, while a Fable limit reset returns only the weekly usage spent on Fable models. The restored allowance also applies in Claude Code, because usage limits are shared across your account.
Run /usage before you hit a wall rather than after. It shows your plan usage bars, and you can press d or w to switch the usage breakdown between the last 24 hours and the last 7 days. The shape of the windows and what counts against them is in how Claude usage limits are measured. When each window resets and what your options are once you have hit one are both worth reading once, so the timing stops surprising you.
How to notice you are being served a different model
Claude can switch models on its own in two situations, and neither one is about running out of allowance.
The first is a safety classifier. On Claude Opus 5 and Claude Opus 5.5, a request that the cybersecurity classifier flags is re-run on Claude Opus 4.8, and Anthropic's help page says "you'll see a notice explaining that the model switched" and "the response will be labeled with the model that answered." Opus 5 does not fall back on biology, chemistry or life-sciences questions; the same page says it "uses similar safeguards for these topics as Opus 4.8". On Opus 5.5, dual-use biology requests and a narrow set of frontier LLM development requests re-run on Opus 5. Attempts to extract the model's internal reasoning are blocked outright on both, with no fallback. On Fable 5.2, Fable 5.1 and Fable 5, a biology flag re-runs on Opus 5 and a cybersecurity flag re-runs on Opus 4.8. In Claude Code, Sonnet 5.5 also falls back to Sonnet 5 for cybersecurity flags; biology flags end in a refusal. You can switch back from the model picker, and you can turn the automatic switch off in Settings under Capabilities, which makes the conversation pause instead.
The second is an availability fallback that you configure yourself. A fallback chain sends the request on to the next model when the first one will not take it:
claude --fallback-model sonnet,haikuThe same chain in settings.json:
{
"fallbackModel": ["claude-sonnet-5", "claude-haiku-4-5"]
}Claude Code tries each entry in order until one accepts the request. That is useful when your first choice is overloaded, which is a separate failure with its own error text: what the overloaded error means and how to clear it. It also means the model answering you is not always the model you selected, so check instead of assuming. Run /model to see which models your account can reach and which one is selected now. In the chat app, read the model label on the response itself.
What a model change costs you mid-task in Claude Code
Switching with /model changes which model answers your next request. The conversation history stays where it is, so you do not lose the session or the files already read. What changes is judgement, and that shows up in the middle of multi-step work.
A lighter model picks up the same context and often takes a different route through it, re-reading files the previous model had already understood, or writing a simpler fix than the plan called for. That is not a fault. It is a smaller model doing its best with a plan it did not write. Each model also keeps its own prompt cache, so the first request after a switch re-reads the whole conversation with no cache hits.
Two habits make this cheap. Switch at a task boundary rather than halfway through a multi-file edit, so the new model starts on a clean step. And use the built-in split when the work suits it: the opusplan setting uses Opus while you are in plan mode and switches to Sonnet automatically for execution, which is the same trade made on purpose instead of by accident.
There is one more reason the Opus allowance disappears faster than your typing suggests. Claude Code sends the whole conversation with every request, and every batch of tool results becomes another request carrying that history. A one-line question in a session you have had open all day still draws usage for the entire conversation. Your first message after a break longer than the prompt cache lifetime also misses the prompt cache, so the cached prefix is rebuilt at the cache-write rate: 1.25 times standard input pricing for five-minute caching or twice standard input pricing for one-hour caching. Clearing between unrelated tasks with /clear reduces the conversation history carried into the next task.
When the API beats a bigger subscription
A subscription is a fixed price with included usage limits; eligible plans occasionally receive limit resets that restore the offered allowance without changing tiers. The API bills by token consumption. The decision point is the shape of your work rather than its volume.
Stay on Max while the work is interactive and you are the bottleneck. Anthropic's Claude Code cost documentation, checked on 10 November 2026, puts enterprise usage at around 13 USD per developer per active day and 150 to 250 USD per developer per month, with 90% of users below 30 USD per active day. Those are published figures for organisations paying per token, and they land close to what Ultra costs.
Move to the API when the work runs without you: scheduled jobs, batch pipelines, many agents in parallel, or billing that has to be split per project. Claude Code supports repository-level usage attribution through OpenTelemetry on subscription access too, including token-based cost estimates through claude_code.cost.usage; those estimates do not allocate your subscription invoice. Separate API keys and workspaces let you track actual API spending per project. The pricing page, checked on 10 November 2026, also shows monthly API credits on Max and Ultra, 100 USD on Max and 200 USD on Ultra, for building on the Claude Platform.
Usage credits sit between the two, and they carry one property worth knowing before you lean on them. By default, the main conversation's prompt cache lifetime is one hour on a subscription's included usage and drops to five minutes once you are drawing on usage credits, so pauses longer than five minutes cause the cache to go cold and the context to be reprocessed. Claude Code v2.1.242 and later lets you set promptCacheTtl to 1h to keep the one-hour lifetime on usage credits. The full side-by-side with numbers is in the API against subscription cost comparison.
For readers in Taiwan and Hong Kong, add a note on price and a note on timing. Plan prices are listed in USD, and what lands on your card depends on local billing and tax, which is covered in what Claude Max costs in Taiwan and Hong Kong. The reset windows follow your account rather than your office hours, so a weekly reset can fall in the middle of a Tuesday afternoon in Taipei. The limit message states the exact reset time. Read it instead of guessing, and schedule the heaviest Opus work for just after it.
FAQ
Does Claude Max include models that Claude Pro does not?
On Anthropic's pricing page, checked on 10 November 2026, Pro, Max and Ultra all list the same four model families: Fable, Opus, Sonnet and Haiku. The difference between the plans is allowance rather than catalogue, with a few conditions attached. On Pro, Fable runs on usage credits outside the 9 to 16 November 2026 launch week, while Max and Ultra include up to 50% of your weekly usage limits on Fable at no extra cost. The 1 million token context window on Opus 5.5 and Opus 5 comes with every paid plan, Pro included. Model availability changes between releases, so read the plan comparison at claude.com/pricing before you buy rather than trusting a list you read months ago.
What happens in Claude Code when I run out of Opus?
Claude Code prints You've hit your Opus limit along with the time the window resets. It does not move you to another model on its own. The Opus and Sonnet limits apply only to that model family, so running /model and choosing a model outside the family lets you carry on immediately. The plan-wide messages behave differently: You've hit your session limit and You've hit your weekly limit count across all models, and switching model does not help there.
How do I know which model actually answered me?
In the chat app, a response that came from an automatic fallback is labeled with the model that answered, and a notice explains why the switch happened. In Claude Code, run /model to see which models your account can reach and which one is selected, and run /usage for your plan usage bars, pressing d or w to switch the usage breakdown between the last 24 hours and the last 7 days. If you configured a fallback chain with --fallback-model or the fallbackModel setting, remember that the reply can come from the second entry whenever the first is unavailable.
When is the API a better answer than upgrading to Ultra?
Upgrade the subscription while the work is interactive and its included allowance covers your needs. Move to the API when the work runs without you: scheduled jobs, batch pipelines, parallel agents, or billing that has to be split per project. Usage credits are the middle option, with one catch to plan around: the main conversation's prompt cache lifetime drops from one hour to five minutes by default while you draw on them, so pauses longer than five minutes cause cache misses unless you set a longer lifetime.