Side by side · November 2026

Claude Sonnet 5.5 vs Claude Fable 5.2

Claude Sonnet 5.5 scores higher on the Models at Work leaderboard (72 vs 65 out of 100). On independent tests alone they are also level (83 vs 86). Claude Sonnet 5.5 is cheaper: 1,000 support replies cost about $10 at list price, against $40 on Claude Fable 5.2, which is 4 times as much. On the Artificial Analysis Intelligence Index, one of the independent tests, Claude Fable 5.2 leads 65 to 55.

Updated Nov 18, 2026 · every number links to who measured it
Claude Sonnet 5.5 · AnthropicClaude Fable 5.2 · Anthropic
Leaderboard rank#6#14
Score (0–100)7265
Kind of workEveryday defaultHardest problems
Price per 1M tokens$2 to read · $10 to write$8 to read · $40 to write
Reads at once1M tokens (about 750,000 words)1M tokens (about 750,000 words)
Where you can get itClaude Team and Enterprise, Amazon Bedrock, Gemini Enterprise Agent Platform (Google Cloud), Microsoft Foundry, Microsoft 365 Copilot, GitHub CopilotClaude Team and Enterprise, Amazon Bedrock, Gemini Enterprise Agent Platform (Google Cloud), Microsoft Foundry
Can a company host it itselfNoNo
Artificial Analysis Intelligence Index5565
Humanity's Last Exam, Diamond (Scale)37.5072.40
Vals Finance Index64.53% (rank 6)63.58% (rank 7)
Vals Tax Agent Bench41.96% (rank 12)32.09% (rank 29)
Vals Legal Research Bench48.08% (rank 15)46.15% (rank 20)
Humanity's Sixth Sense (Scale)Not measured59.50
Epoch Capabilities IndexNot measured173
Remote Labor Index (Scale)Not measured41.25% of projects automated
Professional Reasoning Benchmark - Finance (Scale)Not measured47.07
Professional Reasoning Benchmark - Legal (Scale)Not measured48.97
Best forSupport replies at volume, Drafting proposals and reports, Summarising meetings and long documents, Answering questions from company documentsLong projects left to run with little supervision, Financial models in spreadsheets, Multi-step research, Work Claude Opus 5.5 gets wrong

Cost of 1,000 jobs

1,000 of theseClaude Sonnet 5.5Claude Fable 5.2
Support repliesthe customer message, the account notes and two help articles read; a 300-word reply written$10$40
Documents processeda two-page invoice or form read; the fields pulled out$7$28
Calls summariseda 40-minute call transcript read; a summary and next steps written$26$104
Contracts revieweda 40-page contract read against a checklist; a memo of issues written$75$300
Drafts writtena brief and one example read; a 600-word email or proposal section written$11$44

Worked out from each vendor’s list price and a fixed size for each job. A product built on a model charges its own price on top. See what every model costs for the same jobs.

Choose Claude Sonnet 5.5 for

  • Support replies at volume
  • Drafting proposals and reports
  • Summarising meetings and long documents
  • Answering questions from company documents

A sound everyday default: within a point of Claude Opus 5.5 on Vals AI's test of finance work (64.5% against 65.3%) at half its list price.

Caveats: The saving depends on the effort setting: at the top setting Artificial Analysis's test run cost $10.02 per task, level with Opus 5.5 ($10.04). On Scale's Humanity's Last Exam, a set of expert-level questions, it scores 37.5 against 55.0 for Opus 5.5: move up for the hardest analysis. In Microsoft 365 Copilot it has been rolling out since 30 September with unpublished usage limits; Claude models are off in the EU, EFTA and UK until an administrator allows Anthropic as a subprocessor (a firm allowed to handle your data). Vals AI notes that some of its test tasks were answered by a fallback model after Sonnet 5.5 refused them. Up from 11th to 6th because Vals AI's finance, tax and legal research tests now count toward the score. Places 6 to 8 share a score.

Choose Claude Fable 5.2 for

  • Long projects left to run with little supervision
  • Financial models in spreadsheets
  • Multi-step research
  • Work Claude Opus 5.5 gets wrong

First on Artificial Analysis's index (65) and on Scale's Remote Labor Index: it finished 41.3% of real freelance projects to a client's standard, Claude Opus 5.5 21.3%. Twice Opus's price, and not first on Vals AI's finance or tax tests.

Caveats: Anthropic's own advice is to start with Opus 5.5 and move up only when its answers fall short. It writes more per answer: Artificial Analysis's test run cost $26.54 per task at the top effort setting against $10.04 for Opus 5.5, a wider gap than the list prices suggest. On Claude Team it is offered on premium seats, up to half the weekly usage limit; Enterprise bills usage at API rates, per Anthropic's pricing page. Down from 3rd to 14th: Vals AI's finance, tax and legal research tests now count, and it gets 63.6%, 32.1% and 46.2% right, behind models a quarter of its price. Only one model here costs more. Places 14 to 19 share a score. Vals AI ranks it first on its analyst-research and Excel-modelling tests (69.5% and 77.4%) but 29th on corporate tax questions (32.1%): have a person check tax work.

What people running them say

Claude Sonnet 5.5: Less discussed than Opus or Haiku because it just works; the news was October's price cut for text it has already been sent once and reads again, to $0.10 per million tokens (the unit vendors bill by).

Claude Fable 5.2: One Hacker News (developer forum) launch thread, read through Signals: a commenter who tried it says it pulls ahead of Opus 5.5 only on very ambitious requests, another says Opus 5.5 at a high effort setting was enough for a software project of 125,000 lines, and a third asks how anyone measures the difference at the price. Too little to score.

Questions people ask

Which is better, Claude Sonnet 5.5 or Claude Fable 5.2?

Claude Sonnet 5.5 scores higher on the Models at Work leaderboard (72 vs 65 out of 100). On independent tests alone they are also level (83 vs 86). Claude Sonnet 5.5 is cheaper: 1,000 support replies cost about $10 at list price, against $40 on Claude Fable 5.2, which is 4 times as much. On the Artificial Analysis Intelligence Index, one of the independent tests, Claude Fable 5.2 leads 65 to 55.

When should I choose Claude Sonnet 5.5 over Claude Fable 5.2?

Choose Claude Sonnet 5.5 for support replies at volume, drafting proposals and reports, summarising meetings and long documents, answering questions from company documents. A sound everyday default: within a point of Claude Opus 5.5 on Vals AI's test of finance work (64.5% against 65.3%) at half its list price.

When should I choose Claude Fable 5.2 over Claude Sonnet 5.5?

Choose Claude Fable 5.2 for long projects left to run with little supervision, financial models in spreadsheets, multi-step research, work claude opus 5.5 gets wrong. First on Artificial Analysis's index (65) and on Scale's Remote Labor Index: it finished 41.3% of real freelance projects to a client's standard, Claude Opus 5.5 21.3%. Twice Opus's price, and not first on Vals AI's finance or tax tests.

Which is cheaper, Claude Sonnet 5.5 or Claude Fable 5.2?

Claude Sonnet 5.5 costs $2 per million tokens it reads (input) and $10 per million tokens it writes (output) on Anthropic's list price (Claude pricing, read Nov 18, 2026). A token is about three-quarters of a word. Claude Fable 5.2 costs $8 per million tokens it reads (input) and $40 per million tokens it writes (output) on Anthropic's list price (Claude pricing, read Nov 18, 2026). A token is about three-quarters of a word.

What do 1,000 jobs cost on Claude Sonnet 5.5 and Claude Fable 5.2?

At list price: 1,000 support replies cost $10 on Claude Sonnet 5.5 and $40 on Claude Fable 5.2; 1,000 documents processed cost $7 on Claude Sonnet 5.5 and $28 on Claude Fable 5.2; 1,000 calls summarised cost $26 on Claude Sonnet 5.5 and $104 on Claude Fable 5.2; 1,000 contracts reviewed cost $75 on Claude Sonnet 5.5 and $300 on Claude Fable 5.2; 1,000 drafts written cost $11 on Claude Sonnet 5.5 and $44 on Claude Fable 5.2. A product built on either model charges its own price on top.

Scores come from the Models at Work leaderboard: Score is 0 to 100 and computed, not typed. 40% is independent testing (Artificial Analysis, Arena, Scale, Epoch, Vals and others; each test is scaled against the best score on this board, and tests of real work, such as finished freelance projects or professional tasks, count double the exams). 30% is price: the vendor's list price for a typical mix of reading and writing, on a fixed scale from $0.05 to $60 per million tokens, so a model that costs a fifth as much gains real ground. 15% is what people report after running it, from Signals and linked threads. 15% is fit: how much text it can take in at once and where a business can get it (the products and clouds that carry it), with a small credit for open weights. A missing component drops out and the rest are reweighted. A model with no independent test yet is provisional and ranks below every measured one. Within three points is a tie. Vendor-published figures are labelled and never counted. Prices are vendor list prices where published; a typical mix assumes three parts reading to one part writing.