Anthropic releases Claude Fable 5 for Public Use

BlockBeats News, June 10th. Anthropic launched the public version of its Mythos model on Tuesday local time, but imposed restrictions preventing its use in high-risk areas such as cybersecurity.

Anthropic stated that the new version, Claude Fable 5, is the strongest model the company has built for more widespread applications, praising its performance in software engineering and data analysis. Anthropic had previously announced that Mythos had discovered thousands of software vulnerabilities. Subsequently, the company restricted the use of this model through the "Glasswing" program to around 200 institutions, including the U.S. government.

Anthropic mentioned that extensive testing has been conducted to ensure that users cannot manipulate Claude Fable 5 to bypass its guidelines and perform restricted operations. Dianne Penn, Head of Product Management, Research, and Lab at Anthropic, stated that based on early customer feedback, although Fable 5 is priced higher, the token consumption required to perform tasks is lower, thereby reducing the overall cost per task.

Anthropic also stated that users who have access to the Claude Mythos preview version (i.e., the version without security protection mechanisms) will be able to upgrade to the all-new Claude Mythos 5. The pricing for these two models is $10 per million input tokens and $50 per million output tokens.

Click the original text link below to join the BlockBeats · Feishu AI News Channel and stay informed about global AI hotspots and news 24/7.