Anthropic releases Claude Opus 5 with performance approaching Fable 5

Anthropic has officially launched Claude Opus 5, a new flagship model designed to serve as a more efficient, everyday assistant. The release replaces the previous Opus 4.8 iteration and is positioned to deliver performance that closely rivals the company's top-tier Fable 5 model while significantly reducing operational costs.

According to Anthropic, the new model is optimized for coding, knowledge work, and business workflows. While Fable 5 remains the company's most powerful offering for specialized, high-intensity tasks such as cybersecurity exploitation and advanced biological research, Opus 5 is intended for a broader user base. The company reports that Opus 5 requires less effort to achieve results similar to those of the more restricted Fable 5.

Pricing for the new model is set at $5 per million input tokens and $25 per million output tokens, matching the cost structure of the outgoing Opus 4.8. Users also have access to a Fast mode, which provides speeds 2.5 times higher than the default setting at double the standard price. Anthropic notes that the new model is expected to trigger safety interventions approximately 85% less frequently than Fable 5.

The model is now available across all platforms and serves as the default option for Claude Max, while remaining the strongest model available within Claude Pro. Anthropic has also released a dedicated prompting guide to assist users in maximizing the capabilities of the new architecture. Integration for enterprise users is supported through Amazon Bedrock, where the model is now available for production inference workloads.