Price gap is five-fold
Claude Fable 5.1 lists at $10/$50 per 1M tokens. Claude Sonnet 5 lists at $2/$10. The input price is five times higher and the output price is five times higher. There is no published Sonnet 5 score on Terminal-Bench 4.0 to anchor a head-to-head, so the comparison has to be on price and stated capability, not benchmark.
| Dimension | Fable 5.1 | Sonnet 5 |
|---|---|---|
| Input $/1M | $10.00 | $2.00 |
| Output $/1M | $50.00 | $10.00 |
| Cache read $/1M | $0.25 | not published |
| Five effort levels | Yes | No |
| Terminal-Bench-Science 0.1 | 52.6% | not published |
| Best for | Agentic science, cache-heavy agents | High-volume production, cost-sensitive workloads |
When to pick Sonnet 5
Routine coding at scale, structured extraction, classification, summarisation, and any workload where the cache hit rate is modest. Sonnet 5 is the right answer for most teams.
Ready to test the workflow?
Create account & add creditsWhen to pick Fable 5.1
Agentic scientific research, life-sciences workflows, Mythos-gated compute, and long-running cache-heavy agents. The five-fold price premium is only worth it on these workloads.
Frequently asked questions
Should I use Claude Sonnet 5 or Claude Fable 5.1?
Sonnet 5 for cost-sensitive production workloads, Fable 5.1 for agentic science and cache-heavy agents. The five-fold price gap means the answer for most teams is Sonnet 5.
Put Claude Fable 5.1 to work
Fund a controlled evaluation, start with a prepared prompt, and measure quality and cost on your own workload.