Anthropic's Claude Opus 5 delivers near-frontier intelligence at half the price

Anthropic has launched Claude Opus 5, a new AI model designed for daily use that brings frontier-level reasoning at half the price of its top-tier Claude Fable 5. On coding benchmarks like Frontier-Bench v0.1, Opus 5 more than doubles the performance of its predecessor Opus 4.8 while maintaining the same cost. It also achieves scores within 0.5% of Fable 5 on CursorBench 3.2 at max effort, but at half the cost per task, making it the most cost-effective option for complex software engineering. Beyond coding, Opus 5 dominates knowledge work evaluations: on ARC-AGI 3, it scores three times higher than the next best model, and on Zapier AutomationBench, its pass rate is 1.5x the competition at the same cost. It also outperforms all models on OSWorld 2.0 computer use benchmarks at any cost point, surpassing Fable 5's best result at just over a third of the cost.

Opus 5 also shows strong agency and thoroughness in real-world tasks. In early access testing, it was given a drawing of a machine part with no direct way to view it—Opus 5 autonomously built a computer vision pipeline to extract the geometry from raw pixels and reconstructed the full 3D model. No competing model could solve the same task in five attempts. It also fixed a real bug in a popular open-source package manager, catching an edge case that the community's patch missed, while a competing model only fixed the surface symptom. An engineer at a trading firm used Opus 5 to build a complete market data feed for a new exchange in a single session, including a test harness to validate the code when no live feed was available. These examples highlight Opus 5's ability to iterate carefully and verify its own work, making it a powerful tool for demanding professional workflows. The model also brings significant improvements to scientific research, especially in organic chemistry and structural biology.