Anthropic released Fable and Mythos 5.1 on Tuesday, introducing updates designed to cut token costs and reduce false-positive safeguard triggers, TechCrunch reported.
The unrestricted model, Fable 5.1, is available immediately through the Anthropic API and on cloud platforms. Mythos 5.1 remains restricted to registered Anthropic partners conducting research in cybersecurity or life sciences.
Privacy and enterprise controls
Enterprise customers will gain access to a zero data retention service called Enterprise Frontier Safeguards this fall. The service lets organizations run Anthropic models on their own infrastructure without data outflows, while giving clients control over misuse monitoring for agents and human users. Anthropic stated that it has never trained on enterprise data without explicit permission and never will.
Benchmarks and safety findings
Both models set benchmark records on Terminal-Bench 4.0 for command-line coding and Humanity’s Last Exam for general reasoning. Before release, the models produced three novel scientific findings, including a custom GPU optimization and a high-resolution map of Venus assembled from existing photographs.
Anthropic’s system card rates Mythos 5.1 as low-risk for automated AI development, noting that its ability to accelerate internal research and development aligns with current trends. In safety evaluations, the system card noted that Mythos 5.1 shows a slight regression on misaligned behavior compared to Opus 5, but improves over Mythos 5 and Claude Sonnet 5. Anthropic reported that the model cooperates with human misuse and unverifiable claims of authorization more readily than Opus 5, but is less likely to ignore explicit constraints, hallucinate inputs, or falsely claim to have completed tasks.
