Anthropic released Claude Opus 5 on Thursday, positioning it as a practical alternative to its previously restricted Fable 5 model. The new system performs comparably to Fable 5 across many domains and shows marked improvement in complex coding tasks. This launch follows weeks of tension between the company and the US government regarding export controls on the more powerful Mythos-class model. That same regulatory pressure coincided with a separate security incident involving OpenAI, which kept industry attention focused on AI safety. By bringing Opus 5 to market, Anthropic aims to offer a compliant version that retains high utility without triggering the same export restrictions. The company explicitly states that Opus 5 comes close to Fable 5’s capabilities while maintaining the necessary safeguards required for international deployment.
- Opus 5 is designed to meet US export control requirements where Fable 5 was previously blocked.
- The model demonstrates specific gains in handling difficult programming assignments.
- Its release immediately follows renewed scrutiny of large language model security practices.
