← Back to all free guides
A free guide by Mariah Brunner ↗ INSTAGRAM ↗ TIKTOK

Fable 5
Is Here

Claude’s most powerful public model. Here’s what it’s best for, when to leave it alone, how to turn it on, and the 5 prompts I’d run on it first.

Anthropic launched Fable 5 in June, and the newest version is now Fable 5.1. Anthropic’s own model page calls it the pick "for demanding reasoning and long-horizon agentic work." (Models overview)

Most powerful doesn’t mean use it for everything. The people who get the most out of Fable know when to reach for it and when to stay on a cheaper model.

How You Get It Now

The free window I filmed about is over. Fable isn’t on the Free plan. On Pro, it runs on pay-as-you-go usage credits. On Max and Ultra, up to 50% of your weekly limit can go to Fable at no extra cost. (Fable on your plan)

When To Use It

01

The hard, long, complicated stuff

Deep research, giant documents, big multi-step projects and complex builds. Anything with a lot of moving parts where you need it to hold the whole thing and not lose the thread halfway through.

The receipt from launch: Stripe reported that Fable 5 "compressed months of engineering into days." In one 50-million-line Ruby codebase, Anthropic says it did a codebase-wide migration in a day that would have taken a whole team over two months by hand. (Anthropic’s launch post) If a task feels too big to hand to AI, that’s the one to give Fable.

02

Not for quick, everyday tasks

Fable is the slowest and priciest model in Claude’s main lineup. On the API it costs 2.5 times what Opus 5.5 does, and it eats your usage faster. For a quick email or a fast question, you’re paying for power you aren’t using. (Pricing)

Keep Opus 5.5 or Sonnet 5.5 for everyday work. Match the model to the size of the job.

How To Turn It On

01

In the Claude app

Click the model name next to the send button and pick Fable 5.1. If you don’t see it, look under More models. You can switch mid-chat, and the change starts with the next reply. (How to change models)

02

In Claude Code

Run /model fable, which points to Fable 5.1. Heads up: that also saves Fable as your default, so switch back with /model opus when the heavy task is done. To make it one project’s default, add "model": "fable" to that project’s .claude/settings.json. (Model docs) New to that file? My power-user settings guide walks through it.

Three Things To Know

01

Thinking is always on

You can’t turn thinking off on Fable 5.1. It decides how hard to think on its own, and the lever you control is Effort, right next to the model name. Pick High or Max for the hardest work. (Effort settings)

02

A 1 million token memory

In chat on paid plans, Fable 5.1 holds about 1 million tokens, so you can feed it huge documents, whole projects or a year of notes in one go. Opus 5.5 and Sonnet 5.5 have 1M now too, so Fable’s edge is how well it reasons over all that, not the size. (Context window sizes)

03

The safety handoff

On some sensitive topics, mainly certain cybersecurity and biology requests, Fable hands the answer to an Opus model. Anthropic said that at Fable 5’s launch, more than 95% of sessions had no handoff at all, and Fable 5.1 triggers it less often. If the style suddenly shifts on a touchy subject, that’s what happened. It’s a safety feature, not a bug. (Fable 5.1 safeguards)

5 Prompts To Run On Fable First

These show off what Fable does best: hold a mountain of context, reason through complexity and carry a long job end to end. Switch to Fable 5.1 first, then paste one in and fill in the brackets.

01

Deep research report

Research wide, then synthesize into something you’d pay a consultant for. Turn on web search first so it can cite real sources.

Deep Research Report
You are a senior research analyst. I want a deep, decision-ready report on: [TOPIC, e.g. "the at-home red light therapy market for women 30-50 in the US"] Why I need it: [your real goal, e.g. "I'm deciding whether to launch a product in this space in the next 6 months"] Use web search and cite a source for every fact, number and price. Then do this in order: 1. Map the landscape: the major players, what they sell, how they position themselves, and rough pricing. 2. Find the real demand signals: who is buying, what they complain about, and what they wish existed (reviews, forums, search trends). 3. Name the 3 biggest gaps a new entrant could win. 4. List the top 5 risks and rate each one low, medium or high. 5. Give me a clear call: go, wait, or don't, and the single most important reason why. Format: a 5-bullet executive summary up top that a busy person can read in 30 seconds, then the detail under clear headers. If you can't find a source for something, say so instead of guessing. End with the 3 questions I should answer next.

02

Make sense of a giant document

Attach a long contract, report, transcript or PDF first, then paste this. Because Fable holds the whole thing, it catches the buried clause on page 38 that a skim would miss.

Giant Document
I attached a long document: [what it is, e.g. "a 40-page commercial lease," "a 2-hour meeting transcript," "our full Q2 analytics export"]. Read all of it, start to finish, before you answer. Then give me: 1. The 30-second version: what this document is and the 5 things that matter most in it. 2. The details I'd miss: anything buried, unusual, risky or easy to overlook. Quote the exact line and tell me why it matters. 3. The numbers that count: every important figure, date, deadline or dollar amount in one clean list. 4. What I should actually do: the concrete next steps or decisions this document forces. 5. The questions to ask: what's missing, vague, or worth pushing back on before I sign, act or share this. Cite the page or section for everything you pull. If something is unclear, say so instead of smoothing over it. This is not legal or financial advice, so flag anything I should run past a professional.

03

Plan and build a real project

Hand it the whole build, not a tiny piece. The "ask me 5 questions first" line forces it to understand before it acts. It plans great in the app, and in Claude Code it can actually build the files.

Plan And Build
I want to build: [the real thing, e.g. "a simple booking site for my studio where clients pick a time and pay a deposit"]. What I have: [your starting point, e.g. "nothing yet," "a Squarespace site," or "a folder of files I'll share"]. My skill level: [e.g. "I'm not a developer but I can follow steps"]. Don't start building yet. First: 1. Ask me up to 5 questions you need answered to get this right. Wait for my answers. 2. Then give me the full plan in plain English: the pieces involved, the order to build them, and what each step does. 3. Flag the 2 or 3 spots most likely to trip me up and how we'll handle them. Once I approve the plan, build it one step at a time. After each step, tell me what you did, what I should check, and what's next. Keep every earlier decision in mind the whole way. Never delete or overwrite my existing files without asking first.

04

Turn a year of notes into a plan

Dump the brain dumps, voice-note transcripts and scattered docs in the chat, then paste this under them. The contradictions step is where it earns its keep.

Notes Into A Plan
Below (or attached) is a big pile of my notes: [what it is, e.g. "a year of journal entries," "every idea in my notes app," "all my customer call notes"]. It's unsorted on purpose. I want you to make sense of it. 1. Read all of it before responding. 2. Tell me the 5 to 7 themes that keep showing up, including ones I might not have noticed. Use my own words where you can. 3. Show me the contradictions: where I say I want one thing but my notes point somewhere else. Quote both sides. 4. Pull out the good ideas that got buried, the ones worth acting on. 5. Give me one clear focus for the next 90 days based on what the notes actually show, plus the first 3 actions for this week. Be honest, even if it's a little uncomfortable. I want the truth, not a flattering summary. Only draw from what's in the notes, and tell me when you're reading between the lines.

05

The big complex untangle

The Stripe-migration energy for your own tangled system. Be specific about what "done" looks like. That’s what turns a vague answer into a real plan.

The Big Untangle
I have a complicated system I need untangled: [Describe it, e.g. "my business runs on 6 spreadsheets that don't talk to each other," "our onboarding process has 22 steps and nobody knows why," or "3 years of finances across 4 tools"] What "done" looks like for me: [e.g. "one clean system I understand and can keep up with"] I'll paste or attach what I have below. Work through it like this: 1. Map the current setup back to me in simple terms so I know you get it. Where are the overlaps, gaps and broken handoffs? 2. Show me the clean version: what this should look like when it's done. 3. Give me a migration plan in phases I can do without breaking what's running today. Lowest-risk steps first. 4. For each phase, tell me exactly what to do and how I'll know it worked. 5. Call out the one thing most likely to go wrong and the safety net for it. Ask me questions before step 3 if anything is unclear. I'd rather you think hard once than give me a shallow answer fast.

The Real Win

Reach for Fable when a task is too big, too long or too complex for anything else, and leave it alone for the quick stuff. Want to stretch your usage further? Try my Fable usage limit fix.