Claude Opus 5 Delivers Fable 5 Performance at 50% the Cost

Anthropic’s Claude Opus 5 is said to match Fable 5-level performance at about half the cost.

It posts strong results in reasoning, coding, agentic search, and broader knowledge work.

The model keeps the same token pricing as Claude Opus 4.8, improving cost efficiency for developers and businesses.

Its main limits are weaker long-form reasoning and cybersecurity performance, which may narrow some use cases.