Anthropic released Claude Opus 5 today. It says Opus 5 comes close to Claude Fable 5 at half the price. For users, that shifts the choice from pure capability to how much work they can get done for each task unit.
Claude Max and Claude Pro
Opus 5 is now the default model on Claude Max. It is the strongest model on Claude Pro. That puts the new release directly into the settings many users will encounter first, instead of leaving it as a niche option.
Frontier-Bench and GDPval-AA
Anthropic says Opus 5 is the new state-of-the-art on coding and knowledge work evaluations like Frontier-Bench and GDPval-AA. It also says Opus 5 surpasses all other models on Frontier-Bench v0.1. On CursorBench 3.2, at max effort, it lands within 0.5% of Fable 5's peak score at half the cost per task.
Mythos 5 and safety
The same release carries a sharper limit. Opus 5 remains behind Mythos 5 on cybersecurity tasks, even after Anthropic says it has improved substantially on cyber work as a byproduct of broader capability. The company also says its automated behavioral audit found Opus 5 to be its most aligned model to date, with the lowest rates of deceptive behavior and the least susceptibility to misuse.
The practical question now is not whether Opus 5 exists. It is already available today. The open question is how Anthropic will price access beyond Claude Max and Claude Pro, since the release only spells out the default and strongest-model placements for those two tiers.
