Anthropic Launches Claude Fable 5.2 and Mythos 5.2
Anthropic releases Claude Fable 5.2 and Mythos 5.2, featuring the same underlying model with two safeguard levels and scoring 49.1% on FrontierCode 2.0.
Stock photo for illustration only, not from the actual event
- Anthropic introduces Claude Fable 5.2 and Mythos 5.2 as a single model with two distinct safeguard tiers.
- Fable 5.2 achieves a 49.1% score on the FrontierCode 2.0 benchmark.
- Pricing is set at $8 per million input tokens and $40 per million output tokens.
- Comes equipped with a massive 1-million-token context window.
Artificial intelligence company Anthropic has announced the release of its latest models, Claude Fable 5.2 and Claude Mythos 5.2. Despite the distinct names, they share the exact same underlying model architecture but are deployed with two different levels of safety safeguards.
In terms of benchmark performance, the Fable 5.2 model secures a 49.1% score on FrontierCode 2.0, demonstrating competitive capabilities in coding and technical task execution.
Stock photo for illustration only, not from the actual event
Furthermore, the model architecture supports a 1-million-token context window, enabling users to input extensive documents and large codebases into the system simultaneously without losing contextual awareness.
Releasing a single foundational model configured with split safeguard levels highlights Anthropic's strategic effort to cater to both flexible developer needs and strict enterprise compliance standards. Maintaining one core architecture while varying safety guardrails allows the company to streamline training costs and mitigate AI safety risks effectively across different deployment environments.
Regarding pricing structure, the models are billed at 8 dollars per million input tokens and 40 dollars per million output tokens, aligning with the industry standards for top-tier frontier artificial intelligence systems.
Source: MarkTechPost
Found something wrong in this article? Report an issue with this article
Comments
Leave a Comment