Claude review (October 2026): Opus 5.5, Sonnet 5.5 and the new Fable 5.1
Anthropic refreshed every Claude model this autumn. What each plan gets you, where Claude beats ChatGPT and Gemini, where it falls short, and who should pay.
We may earn a commission if you sign up through links on this page. It costs you nothing and never decides what we recommend. Learn more
How we reviewed this: this is a research review. It is built from Anthropic's own documentation, pricing and announcements plus published benchmarks, all listed under Sources and checked on Oct 28, 2026. Scores reflect what each plan offers on paper. When we finish our own hands-on tests, we'll update this page. How we review
Our verdict
Claude by Anthropic
The best pick for long documents and careful writing, with a 1 million token context window across the new 5.5 models and a clear choice over training. Pro at $20 suits most people. Max is for heavy users and anyone who wants Fable 5.1 included.
- Best for
- Writers, analysts, lawyers, researchers and developers who work with long documents, reports, contracts or code.
- Skip it if
- Anyone under 18, and people who mainly want an assistant built into the Google or Microsoft apps they already pay for.
- Pricing
- Free; Pro $20/mo ($17/mo billed yearly); Max 5x $100/mo; Max 20x $200/mo.
Our research scores
What we like
- 1 million token context window on Opus, Sonnet and Haiku 5.5
- Three new models in three weeks, all cheaper or faster than before
- You decide whether your chats are used for training
- Pro is cheaper billed yearly, at $17 a month
What could be better
- 18+ only
- Fable 5.1 isn't part of Pro's usage; you pay for it with usage credits
- Heavy days can hit Pro's usage limits; more room means Max at $100 or $200
- Several model tiers to understand if you build with the API
What’s new in Claude this autumn
Anthropic shipped four new models between September 1 and October 7, 2026:
| Model | Released | What it’s for | API price (in / out, per 1M tokens) |
|---|---|---|---|
| Fable 5.1 | Sep 1 | The top tier: long agent runs, deep research, spreadsheets and slides | $10 / $50 |
| Opus 5.5 | Sep 22 | The most capable everyday model | $4 / $20 |
| Sonnet 5.5 | Sep 28 | The balanced default for most work | $2 / $10 |
| Haiku 5.5 | Oct 7 | Fast and cheap | $0.10 / $0.50 |
Two things stand out. Every 5.5 model reads up to 1 million tokens at once, around 1,500 pages of text, so you can hand Claude a whole contract bundle or a year of reports. And Opus got cheaper: Opus 5.5 costs $4 / $20 in the API, down from $5 / $25 for Opus 5.
Fable 5.1 keeps Fable 5’s price but cuts the cost of cached input to a quarter, which Anthropic estimates makes typical bills about 25% lower.
Which Claude plan should you pick?
| Plan | Price | What you get | Who it’s for |
|---|---|---|---|
| Free | $0 | Claude for everyone, with lower usage | Trying it out, light use |
| Pro | $20/mo, or $17/mo yearly | More usage and Projects; Fable via usage credits | Most people who use it for work |
| Max 5x | $100/mo | Five times Pro’s usage, Fable included | Daily heavy users |
| Max 20x | $200/mo | Twenty times Pro’s usage, Fable included | People who run Claude all day |
Prices from Anthropic’s pricing page, checked October 28, 2026.
Our take: Pro is right for most people, and paying yearly brings it down to $17 a month. Go to Max only when you regularly run out, or when Fable 5.1 is central to your work.
What Claude does best
Long documents. This is Claude’s home ground. A 1 million token window across Opus, Sonnet and Haiku means you rarely need to split a document. Ask for answers with page numbers and quotes, and Claude tends to show its working, which makes checking easy. Our long-PDF guide picks Claude for exactly this reason.
Writing that sounds like a person. Claude is a favourite for drafts, edits and tone changes. It follows style instructions closely, which matters when you want emails or reports in your own voice.
Coding and agents. Opus 5.5 and Fable 5.1 are built for long coding and agent sessions. Anthropic’s launch partners report Fable 5.1 finishing more browser-agent tasks than Opus 5 and Fable 5 while using fewer tokens, though those are the partners’ own tests.
Where it falls short
- 18+ only. Students under 18 need a different tool.
- Fable costs extra on Pro. It’s paid with usage credits unless you’re on Max.
- Usage limits. On busy days, Pro users can hit the cap. Max raises it, at five or ten times the price.
- Fewer built-in extras. ChatGPT and Gemini bundle more side features into one app. Claude is more focused on reading, writing, analysis and code.
Privacy and safety
Claude gives you a clear choice: on Free, Pro and Max, you decide whether your chats can be used for training. Data from Team, Enterprise and the API isn’t used for training at all. That’s the most straightforward default of the three big assistants.
Claude vs ChatGPT vs Gemini
Claude wins on long documents and privacy choice. ChatGPT wins on breadth of features. Gemini wins if your life runs on Google apps. See the full comparison table in our ChatGPT review, and our Gemini review.
Bottom line
If your work involves long documents or careful writing, Claude is the assistant to beat in late 2026, and the 5.5 models made it faster and cheaper. Start on Pro, billed yearly if you’re sure. Move to Max when you hit the limits or need Fable 5.1 every day.
Alternatives to Claude
ChatGPT
by OpenAI
General-purpose assistant for writing, analysis, files, images and voice.
Free; Go $8/mo (US); Plus $20/mo; Pro from $100/mo
Gemini
by Google
Google's assistant, built into Gmail, Docs, Sheets and Search.
Free; AI Plus $4.99/mo; AI Pro $19.99/mo; AI Ultra from $99.99/mo (US)
Privacy & risks
- Conversations can be read by human reviewers. Activity auto-deletes after 18 months by default (you can change this); temporary chats are kept for up to 72 hours.
- A proposed class action over Gemini data tracking (Thele v. Google) was dismissed with leave to amend in July 2026. Google moved to dismiss the amended complaint in September 2026.
Perplexity
by Perplexity AI
Answer engine that searches the web and cites its sources.
Privacy & risks
- Trains on your data
- Yes, opt-out
- Minimum age
- 13+ (minors need a parent's acceptance)
- "AI Data Retention" (use of your data for training) is on by default for Free, Pro and Max. Enterprise data is never used for training.
- In August 2026 the Ninth Circuit vacated a preliminary injunction Amazon had won against Perplexity's AI agent shopping on its site.
Questions people ask
Is Claude free?
Yes, there is a free plan. Pro ($20 a month, or $17 a month billed yearly) adds more usage and features, and Max ($100 or $200 a month) adds five or twenty times Pro's usage.
What is the difference between Opus, Sonnet, Haiku and Fable?
They are sizes. Haiku is the fastest and cheapest, Sonnet is the balanced everyday model, Opus is the most capable mainstream model, and Fable is the top tier for long, hard jobs. In the API, Sonnet 5.5 costs half of Opus 5.5.
Does Claude train on my chats?
On Free, Pro and Max you choose whether your chats can be used for training. Team, Enterprise and API data is not used for training.
Can I use Claude Fable 5.1 on the Pro plan?
Yes, but it isn't covered by Pro's normal usage limits. You pay for it with usage credits. Max plans include it.
Sources
- Claude plans and pricing (Anthropic)
- Introducing Claude Fable 5.1 and Claude Mythos 5.1 (Anthropic)
- Claude Fable models on your plan (Claude Help Center)
- Claude Fable 5.1 overview (Claude Platform docs)
- Claude Opus 5.5 (Anthropic)
- Claude Sonnet 5.5 (Anthropic)
- Claude Haiku 5.5 (Anthropic)
- Consumer terms (Anthropic)