🔍 Read the full analysis: Is Claude Fable 5.2 Worth It? Per-Task Costs By Effort Level on ThorstenMeyerAI.com
Get business pricing on office and shipping supplies
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
TL;DR
Artificial Analysis’s November 10 testing puts Claude Fable 5.2’s per-task cost between $3.10 at Low effort and $26.54 at Max, with scores ranging from 54 to 65. High scored 62 for $9.08, while Max scored 65 for $26.54; those figures are averages from the benchmark, not a fixed price for every real-world task.
Claude Fable 5.2 costs an average of $3.10 to $26.54 per task across five effort settings in Artificial Analysis’s Intelligence Index test, with scores rising from 54 to 65. The figures, accessed November 10, 2026, give businesses a benchmark for weighing additional model reasoning against higher API bills and longer response times; they do not establish what any individual task will cost.
Artificial Analysis tested Fable 5.2 at Low, Medium, High, Xhigh and Max effort across its Intelligence Index, which covers nine evaluations in areas including agentic coding, business documents, science and general knowledge. Its reported average cost per task rose from $3.10 at Low to $3.80 at Medium, $9.08 at High, $14.83 at Xhigh and $26.54 at Max. The corresponding displayed scores were 54, 56, 62, 62 and 65.
The test also found that time per task increased with effort: 6.5 minutes at Low, 8.3 at Medium, 17.9 at High, 26.4 at Xhigh and 46.4 at Max. Artificial Analysis reports output-token use for the full index rising from 34.9 million at Low to 275 million at Max. Those totals describe the benchmark run, not an individual task. The cost-per-task measure is the average API cost of one task in that test.
The step from Medium to High brought a six-point score increase, from 56 to 62, alongside a cost rise of $5.28 per task. Xhigh cost about 63% more than High while both displayed a score of 62. Max added three points over High but cost nearly three times as much. In one separate Artificial Analysis coding evaluation, FrontierCode 2.0, Fable scored 49.1% at High and 48.8% at Max, indicating that the higher setting did not score better on that test.
Is Claude Fable 5.2 Worth It? Per-Task Costs by Effort Level
Artificial Analysis’s November 10 benchmark shows higher effort brought higher average costs, longer task times and scores from 54 to 65. These are benchmark averages, not guaranteed prices or outcomes for real-world tasks.
Five effort settings, rising costs
The Intelligence Index scores rose overall as average cost and time per task increased.
| Effort | Cost per task | Score | Time per task |
|---|---|---|---|
| Low | $3.10 | 54 | 6.5 min |
| Medium | $3.80 | 56 | 8.3 min |
| High | $9.08 | 62 | 17.9 min |
| Xhigh | $14.83 | 62 | 26.4 min |
| Max | $26.54 | 65 | 46.4 min |
Where the trade-off changes
Higher settings do not guarantee proportionally better benchmark results.
- Score rose from 56 to 62.
- Average cost increased by $5.28 per task, from $3.80 to $9.08.
- Score rose by three points, from 62 to 65.
- Max cost nearly three times as much: $26.54 versus $9.08.
Benchmark context
The Intelligence Index covers nine evaluations, including agentic coding, business documents, science and general knowledge.
Same published API rates
Anthropic lists standard rates of $8 per million input tokens and $40 per million output tokens. Benchmark task costs vary with token consumption.
Coding results can differ
In a separate FrontierCode 2.0 evaluation, High scored 49.1% and Max 48.8%, so Max did not score better on that test.
How teams can assess value
Use the benchmark as a reference, then measure results on work your team actually needs to do.
Test the kinds of work for which your team is considering Fable.
Record quality, token use, API cost and waiting time.
Decide whether a higher setting improves the result enough to justify its extra cost and time.
How Effort Changes the Benchmark Bill
The results put a concrete price range around a model setting that can otherwise be hard for teams to budget. Max effort was the highest-scoring configuration in this index, but its reported cost and average completion time were also the highest. For tasks where latency or volume matters, a lower setting could be more practical; for a smaller number of demanding tasks, a higher setting may be worth testing.
These measurements should not be treated as a universal price list or proof that extra effort improves every output. Actual API bills depend on token use and task characteristics, while the reported scores are rounded benchmark results. The comparison is useful as a controlled reference point, not a guarantee of performance or cost in a company’s own workflow.
Products Worth Considering
Apple Mac mini Desktop Computer with M2 Pro, 16GB Memory, 512GB SSD, Gigabit Ethernet
Apple MacBook Air 13-inch Laptop with M5 Chip, 16GB Memory, 512GB SSD, Liquid Retina Display, and 12MP Center Stage Camera (2026, Silver)
Dell UltraSharp U2724DE 27-Inch WQHD IPS Black Monitor
As an affiliate, we earn on qualifying purchases.
Anthropic’s New Model and Pricing
Anthropic announced Claude Fable 5.2 and Claude Mythos 5.2 on November 9, replacing their 5.1 counterparts. The source report says Anthropic lists standard API pricing for Fable at $8 per million input tokens and $40 per million output tokens at each effort setting. The benchmark’s differing task costs reflect varying token consumption, rather than a different published per-token rate for each setting. Anthropic also offers an optional fast mode at $16 per million input tokens and $80 per million output tokens.
Fable is available to users with safeguards for certain high-risk cybersecurity and biology requests, according to Anthropic. Mythos is described as the same underlying model with those safeguards removed, restricted to organizations accepted into Anthropic’s Cyber or Life Sciences Verification Programs. Artificial Analysis’s effort-level table concerns Fable 5.2; it does not establish equivalent benchmark costs for Mythos.
Anthropic recommends starting with Claude Opus 5.5 for most coding and knowledge work and moving to Fable when a task calls for it, according to the source material. In Artificial Analysis’s comparison, Opus 5.5 at Max scored 59 at $10.04 per task, while Fable 5.2 at High scored 62 at $9.08. This is a benchmark comparison, not an independent assessment of every product use case.
Limits of the Cost Comparison
The published figures do not show what Fable 5.2 will cost on a particular company’s prompts, workloads or volume. The source material does not provide a distribution of individual task costs, and the benchmark’s displayed scores are rounded. It is also unclear how closely the nine evaluations reflect the mix of work a given team needs to complete.
The source report describes Anthropic’s claims about model speed, safety testing and prerelease achievements, but those statements are not independently established by the Artificial Analysis cost table. The benchmark results alone do not verify those claims or show that every higher-effort response is more accurate. The reported coding result, where High slightly exceeded Max, further cautions against assuming that more effort always produces a better result.
Testing Effort Settings in Practice
Teams considering Fable can compare effort settings on representative tasks, recording quality, token use, cost and waiting time rather than relying on a single benchmark average. A sensible evaluation would include the tasks for which a stronger model is being considered and check whether a higher setting changes the result enough to justify its added expense.
Artificial Analysis’s figures were accessed November 10, 2026, and the source material does not announce a further benchmark update or a specific next release date. For now, the available evidence is the reported Intelligence Index comparison and Anthropic’s stated pricing and usage guidance; real-world costs and outcomes remain dependent on how the model is used.
Key Questions
What does Claude Fable 5.2 cost per task?
In Artificial Analysis’s Intelligence Index test, average cost ranged from $3.10 at Low effort to $26.54 at Max. These are benchmark averages, not fixed charges for every task.
Which Fable 5.2 effort level had the highest score?
Max effort scored 65 in the reported index results, the highest of the five tested settings. It averaged $26.54 and 46.4 minutes per task in that benchmark.
Is High effort a better value than Max?
In this index, High scored 62 for an average $9.08 per task, while Max scored 65 for $26.54. Whether that trade-off is worthwhile depends on the task; in a separate coding test cited in the source, High scored slightly above Max.
Does every Fable 5.2 task cost the same amount?
No. The reported figures are averages from Artificial Analysis’s benchmark. Actual API costs vary with token use and the task. Anthropic’s stated standard rates are $8 per million input tokens and $40 per million output tokens.
How is Mythos 5.2 different from Fable 5.2?
The source material describes Mythos as the same underlying model without some of Fable’s safeguards, available only to organizations accepted into Anthropic’s verification programs for cybersecurity or life sciences. The effort-level cost results cited here are for Fable.
Source: ThorstenMeyerAI.com
