Anthropic Unveils Claude Opus 5: A Game-Changer in AI Technology
Anthropic has recently launched Claude Opus 5, its newest frontier AI model designed for diverse applications, including coding, research, and business tasks. This latest release reportedly offers a substantial performance boost over its predecessor, Claude Opus 4.8, while maintaining the same API pricing structure. Remarkably, Anthropic claims that Claude Opus 5 approaches the performance of Claude Fable 5 in various coding and productivity tests, but at a significantly lower cost per task.
Claude Opus 5 is made accessible through all of Claude’s applications and APIs, assuming the position of the default model for subscribers of Claude Max. It is also the most robust option available for Claude Pro users, enhancing the functionalities available to subscribers.
How Close is Claude Opus 5 to Fable 5?
When examined using the CursorBench 3.2, a benchmarking tool that tests coding agents on actual software development tasks, Claude Opus 5 achieved a score within 0.5% of Fable 5’s highest performance at maximum effort. Notably, Anthropic emphasizes that this impressive result was achieved at only half the cost per task compared to Fable 5.
Claude Opus 5 OSWorld 2.0 benchmark Anthropic
Claude Opus 5 CursorBench benchmark Anthropic
In a further testament to its capabilities, Claude Opus 5 surpassed Fable 5’s results on the OSWorld 2.0 benchmark, which evaluates how effectively AI agents can operate across applications and complete various tasks, achieving this at just over a third of the cost.
For Pro and standard Team subscribers, Opus 5 could bridge the functionality gap left by Fable 5, particularly after Fable 5 transitioned to a pay-as-you-go credit system on July 20.
How Much Better is Opus 5 Compared to Opus 4.8?
Anthropic asserts that Claude Opus 5 has more than doubled the score of Opus 4.8 on the Frontier-Bench, a benchmarking tool used to assess AI agents on complicated software engineering tasks, all while executing those tasks at a lower average cost.
Claude Opus 5 Frontier-Bench v0.1 benchmark Anthropic
In addition to its scoring improvements, the company notes that Opus 5 shows enhanced capabilities in reviewing its own work, diagnosing bugs, and persevere through challenging tasks instead of halting after initial fixes. For instance, Opus 5 successfully identified an edge case that had been overlooked by an existing community patch.
While these findings stem from Anthropic’s own assessments, actual performance may differ across diverse workloads. Claude Opus 5 retains its cost of $5 per million input tokens and $25 per million output tokens, unchanged from its predecessor Opus 4.8, making it an attractive option for various user needs.
For more information, you can read more Here.
Image Credit: www.digitaltrends.com
