Which Claude model you actually get now depends on your plan

For six weeks this summer, the answer to "which Claude model should I use" kept changing. In late July it settled into a structure that has held ever since, and it has a twist that no amount of reading about model quality will tell you.

You don't just pick a model anymore. You pick a plan, and the plan sets your ceiling.

The model names have moved on since then. Anthropic released Fable 5.1 on September 1 and Opus 5.5 on September 22, followed by Sonnet 5.5 and Haiku 5.5. The structure stayed put: Max includes Fable, Pro reaches it only through usage credits, and Free doesn't get it. So the model at the top of your picker may not be the one at the top of a colleague's, and the difference comes down to what you each pay per month, not anything about your work.

Here's the current state, then what it means for a consulting practice, which is the more useful question.

The state, as of today

Current as of October 10, 2026. Check Anthropic's own plan page and pricing before acting on this. They're the authority; this is our dated summary of them.

Your plan	Strongest model included	Fable 5.1
Free	Sonnet	Not available
Pro	Opus 5.5	Not included; reachable with pay-as-you-go usage credits
Max	Fable 5.1, with Opus 5.5 alongside it	Included for up to half your weekly usage; usage credits after that

Team and Enterprise follow the same split by seat type: premium seats work like Max, standard seats like Pro. Ask whoever runs your workspace.

Anthropic says Opus 5.5 "performs at the level of Claude Fable 5.1 on most work." The current models also have an adjustable effort setting in the model menu, from Low up to Max, and day to day that matters more than the plan question does.

That's the factual picture. Everything below is judgment.

The mistake this invites

The obvious reading is that the good model costs more, so you should pay more.

For a lot of solo consultants that reading is wrong, and the reason is worth being concrete about. Getting Fable 5.1 as a standard part of your plan means Max, which starts at $100 a month against Pro's $20. The real comparison isn't $20 against $100, though. It's $80 a month against what that $80 changes about your work.

At a consulting rate of $150 an hour, $80 is about thirty-two minutes. So the question is whether having Fable 5.1 rather than Opus 5.5 at the top of your picker saves you more than half an hour a month.

For a practice whose AI work is proposals, client updates, case studies, meeting debriefs and scoping, we don't think it does. Fable 5.1 is still the stronger model, but Anthropic's own description of Opus 5.5 is that it matches Fable 5.1 on most work, and drafting a Friday update from a week of messy notes is well inside "most work." The thing that limits that kind of work was never model capability. It's whether you handed over a complete brief, and that stays true on every plan.

Where the ceiling does matter

There's real work where the top tier earns its price, and it has a recognizable shape.

Long, single-handoff assembly. A forty-page RFP you need absorbed and answered. A case study assembled from a project folder with a dozen sources. Seven stakeholder interviews compressed into one diagnostic. These are the tasks where the model has to hold the whole job together across many steps, and the middle of a long run is where weaker models drift. This is the long-leash end of the routing rule, and it's the one row in that table where your plan decides for you.

Work you can't easily check. The expensive failure is the one that survives your review because the output looks finished: a proposal with a plausible but wrong price, or a case study citing something the client never said. The more finished a long draft looks, the more you want the strongest model to have made it, and the harder you should read it anyway.

Volume at the top of the range. If you run several long handoffs a week rather than one a month, paying per use stops being the cheaper option, and a plan that includes the top tier starts to make sense on arithmetic alone.

If none of those describe your month, the model your plan includes isn't a compromise. It's the right tool, and the $80 is better spent elsewhere.

The setting that changes your week

The control more likely to affect your work tomorrow is the effort setting. You choose how much work the model does before it answers, from Low through Medium, High and Extra high to Max.

It lines up with the routing rule almost exactly. For short-leash, low-stakes work you read every word of anyway, keep effort low and take the speed. For a long handoff where a subtle mistake would be expensive, turn it up and let the model check itself before you see the draft.

The practical effect is that the gap between the strong default and the top tier has narrowed for most consulting work. A well-briefed Opus 5.5 run at high effort covers most of the deliverable work that used to call for the tier above, which is why the plan question matters less than the headlines suggest. We've written up how to set effort task by task.

What to do this week

Open your model picker and read it. Your own picker, not an article. It reflects your account, and it's the only reliable record of what you can actually reach. Whatever sits at the top is what every routing recommendation means by "your top tier."

Run one comparison before changing anything. Take a deliverable you produce regularly. Same inputs, same brief, run on your current top model at high effort. Read the output cold and ask whether a stronger model would have changed the result or just the adjectives. That's evidence from your own practice, and it beats anyone's benchmark table, including the one above.

If you want a ready-made deliverable to test with, our free Friday Update Brief is a complete handoff: the brief, a sample week of messy notes, and the reference output to compare against. It costs an email address, and because both runs work from identical inputs, the comparison is fair.

Don't reorganize your practice around a plan. The structure has now held since July, through four new model releases, and that's a good reason to plan around it. It's not a reason to rebuild your workflow every time a new model appears. Try new models eagerly and change your setup slowly.

The part that didn't change

Every model in that table reads the same context files, follows the same standards, and runs the same briefs. The setup that makes AI useful for your practice in particular (who you serve, what you sell and at what floor, how you sound, what's true of each client right now) doesn't depend on the model or the plan. It was worth building in June, and it has carried through every change of terms and every new model since. We've made that argument at length.

If you'd rather start from finished files than a blank page, the Claude Project Kit ships that standing setup: the instructions block and the four practice files. It works the same on whichever model your plan puts at the top of your picker.