Anthropic unveils Fable and Mythos 5.1 with cheaper tokens and fewer restrictions
Anthropic has released its newest AI models, Fable and Mythos 5.1, featuring performance gains, reduced token costs and fewer false-positive safety restrictions. Mythos 5.1 remains exclusive to registered partners in cybersecurity and life sciences.
AI company Anthropic announced on Tuesday the release of Fable and Mythos 5.1, two paired versions of its most advanced AI model. Alongside performance upgrades, the new Fable release includes changes designed to lower token costs and reduce false-positive restrictions caused by the model's safeguards. As with the previous Mythos model, Mythos 5.1 is available only to registered Anthropic partners working in cybersecurity or life sciences research. Fable 5.1, the unrestricted version, is available starting today on cloud platforms or through the Anthropic API.
One of the most significant changes is Anthropic’s previously reported adoption of zero data retention, enabling clients to run Anthropic models on their own infrastructure without data outflows. Previously unavailable for Fable due to security concerns, the high-privacy service Enterprise Frontier Safeguards is expected to roll out in June. The system will still monitor for misuse by agents or human users, but clients will control how monitoring is carried out.
Anthropic also assured customers that their data had not been inappropriately accessed. “Anthropic has never trained on enterprise data without explicit permission, and never will,” the announcement states.
As is typical for an Anthropic release, the new models set records across several benchmarks, including Terminal-Bench 4.0 for CLI-based coding and Humanity’s Last Exam for general reasoning. Anthropic also shared three scientific findings produced by the models before their release, including a custom GPU optimization and a high-resolution map of Venus assembled from existing photos.
The models come with a detailed system card describing their capabilities. The card rates Mythos as “low-risk” for concerns related to automated AI development — where an AI improves itself — which some view as a potential trigger for loss of human control. It says the model's ability to accelerate internal AI R&D is in line with current trends. Regarding general misbehavior, Mythos 5.1 is slightly more prone to it than Opus, possibly due to its enhanced capabilities. According to the system card, it represents a slight regression compared with Opus 5, but an improvement over Mythos 5 and Claude Sonnet 5. It cooperates with human misuse and accepts unverifiable claims of authorization somewhat more readily than Opus 5, yet it is less likely to ignore explicit constraints, hallucinate inputs, or falsely claim to have completed tasks than previous models.
