Anthropic has released Claude Opus 5, a model designed to match the performance of Claude Fable 5 at approximately half the price per token while offering more permissive content classifiers. It serves as the new default model on Claude Max and the strongest option for Claude Pro users.

  • Priced at $5/$25, Opus 5 is positioned as a cost-efficient alternative to Fable ($10/$25) for everyday tasks.
  • The model achieves state-of-the-art results on coding and knowledge work benchmarks like Frontier-Bench and GDPval-AA.
  • It demonstrates exceptional resistance to prompt injection attacks, with success rates dropping to near zero when combined with Auto Mode in Claude Code.
  • Opus 5 excels at local reasoning and subagent tasks but is considered less capable than Fable for global thinking or complex autonomous exploits.

The release provides users with a cheaper option for daily workloads and agentic tasks, though it remains distinct from the Mythos-class intelligence offered by Fable.