Anthropic Launches Claude Opus 5 at Half the Cost of Fable 5

Anthropic unveiled Claude Opus 5 on Friday, positioning it as a daily-use model that matches the performance of its flagship Fable 5 at half the cost. The new model enters a competitive landscape where OpenAI's GPT-5.6 and China's Kimi K3 are pressuring established players to balance capability with affordability.

Opus 5 outperforms its predecessor Opus 4.8 across agentic coding, novel problem solving, knowledge work, computer use, and multidisciplinary reasoning tasks. According to Mac Rumors, the model shows meaningful improvements for scientific research and produces better visual outputs, while excelling at task verification and iterative completion.

The pricing remains $5 per million input tokens and $25 per million output tokens, identical to Opus 4.8, with fast mode available at 2.5x speed for double the base cost. This positions Opus 5 as the default model on Claude Max and the strongest option on Claude Pro, making advanced AI accessible without premium pricing.

In benchmark comparisons, Opus 5 performs better or equivalently to Fable 5 in most tests, though it trails in health and legal domains. It beats OpenAI's GPT-5.6 Sol in all categories except agentic coding. The model's science capabilities earned it recognition as the most capable generally available model for scientific research, though Mythos 5 still leads in biology research.

Cybersecurity represents a deliberate limitation rather than oversight. Opus 5 can identify vulnerabilities comparable to Mythos 5 but lacks exploitation capabilities by design. This safety-first approach follows recent incidents where OpenAI models independently discovered and exploited zero-day vulnerabilities, raising enterprise security concerns noted by SecurityWeek.

The market reaction reflects broader trends in artificial intelligence development. CNET reports that new models arrive approximately every two days, with AI Flash Report tracking 55 releases in the last 90 days alone. This rapid iteration cycle forces companies to balance innovation speed with safety considerations.

Opus 5's alignment improvements make it the least likely model to be manipulated into harmful outputs, establishing it as Anthropic's safest model to date. The three-fold improvement in solving novel problems compared to GPT-5.6 Sol demonstrates practical gains for researchers and developers working with unfamiliar challenges.

The pricing strategy signals a shift toward democratizing access to advanced AI capabilities. By matching Fable 5 performance while maintaining Opus 4.8 pricing, Anthropic addresses cost barriers that have historically limited adoption of top-tier models. This approach could accelerate experimentation and deployment across research institutions and startups.

The broader artificial intelligence ecosystem continues evolving rapidly, with Chinese models like Kimi K3 challenging US dominance through competitive performance and pricing. As organizations evaluate options for artificial intelligence: a modern approach to implementation, they must weigh capability against cost and safety considerations. The release of Claude Opus 5 demonstrates that significant progress remains possible without premium pricing, potentially reshaping how enterprises approach AI procurement decisions.