Claude Fable 5
Claude Fable 5
The best answer money can buy, and it is a lot of money.
Updated
The read
Anthropic's flagship sits at the top of almost every quality benchmark that still discriminates between frontier models, and it is the only one people trust to run for hours without a human reading every step. It is also the most expensive model in this index by a clear margin, and it is not fast. Reach for it when the cost of a wrong answer is higher than the cost of the tokens, and reach for something else when it is not.
Where it shines
- Holds a long, messy task together for hours without drifting
- Leads the field on software engineering benchmarks that are not yet saturated
- Refuses cleanly and explains itself rather than inventing an answer
Worth knowing first
- By far the highest price per token of anything ranked here
- Slow enough that it is the wrong default for interactive work
Index score
89.2
Provisional. This score is read from the product's public capability, not from the one-prompt rebuild the fully reviewed entries went through, so treat it as a placing rather than a verdict.
- Rank
- 01 of 53
- Builds
- LLMs
- Category
- Frontier
- Output
- Text, code and tool calls over an API
- Pricing
- Paid only
- Best for
- Long autonomous runs where being wrong is expensive
Scores are our own editorial judgement, weighted craft 55, speed 10, control 15 and value 20.
Compare it with
Also worth a look
Full index- 03Claude Opus 5Claude Opus 5ProvisionalMost of the flagship, at half the price. The one to reach for by default.Frontier modelPaid only88.5
- 08Gemini 3.6 FlashGemini 3.6 FlashProvisionalFrontier answers at Flash speed, with a free tier you can actually build on.Frontier modelFree tier86.6
- 09Grok 4.5Grok 4.5ProvisionalStrong, current, and unusually willing to answer the question you actually asked.Frontier modelPaid only85.0
