Tariff Tracker: Claude Opus 5 Debuts as Cost-Efficient AI Model for Businesses, Rivals Higher-Priced Leaders on Coding Tasks

What happened: Anthropic announced Claude Opus 5 on July 24, 2026. The model offers frontier-level performance on coding and knowledge work benchmarks - specifically Frontier-Bench v0.1 and GDPval-AA - while priced at roughly 50% less than Claude Fable 5. It also achieves state-of-the-art results on ARC-AGI 3 (tripling the next-best model) and Zapier AutomationBench (1.5x the pass rate of competitors at the same cost). Opus 5 is available as the default on Claude Max and as the strongest model on Claude Pro.

Who: Anthropic, the AI company behind the Claude family of models. Early-access testers included Scott Wu of Devin, Sualeh Asif of Cursor, Wade Foster of Zapier, Alfredo Andere of Basepair (genomics), and Fabian Hedin of Lovable.

Tickers/sectors: No public company tickers were provided in the facts. However, the sectors most impacted include enterprise software, cloud computing, coding tools (e.g., Cursor, Devin), automation platforms (e.g., Zapier), and AI infrastructure. Private companies mentioned are Lovable, Basepair, and Anthropic itself.

Winners/losers: Winners are businesses using AI for software engineering, data analysis, and business process automation - they gain near-frontier intelligence at reduced cost. Companies developing or integrating AI agents for complex coding tasks also benefit. Losers are higher-priced models that cannot match Opus 5's performance-to-cost ratio, particularly for day-to-day operations. No clear losers were named in the facts.

What to watch: Adoption rates among developers and enterprises on platforms like Cursor, Devin, and Zapier. The model's performance on tasks like bug fixing, market data feed building, and scientific research will be key indicators. Also watch for how competitors respond in pricing and benchmark performance.