The launch of Claude Fable 5 marks a significant advancement in AI models, being deemed ready for public use. Anthropic has positioned this model as a “Mythos-class” AI that surpasses all previously released models in terms of capabilities. It excels across various benchmarks, particularly in areas such as software engineering and scientific research.
Initially, Fable 5 will be accessible to users on Pro, Max, Team, and seat-based Enterprise plans until June 22. Following this period, the model will transition to a usage credit system, with plans to restore regular access based on capacity. Meanwhile, Claude Mythos 5, designed for a select group of cyberdefenders, will feature fewer restrictions and is set to be deployed via Project Glasswing in collaboration with the US government.
Concerns previously associated with Mythos centered around cybersecurity threats, but Anthropic asserts that Fable 5 comes with implemented safeguards. If discussions arise about prohibited topics, another model, Claude Opus 4.8, will intervene. The company anticipates that both Fable 5 and Mythos 5 can yield significant benefits in various fields, particularly in enhancing coding and cybersecurity capabilities.