Anthropic has released Claude Opus 5, a new model the company positions as close to its top-end Fable 5 system in capability while costing roughly half as much to run.
According to startupfortune, the model launched on July 24, 2026, arriving with about 26% lower token usage — meaning it burns through less of the billable text that AI systems consume per task — and benchmark results that beat OpenAI's GPT-5.6 Sol. The-decoder likewise reports Opus 5 costs well below Fable 5 while matching or beating it across most benchmarks.
Moneycontrol says Anthropic claims gains in coding, reasoning, scientific research and long-running agentic workflows, alongside new API features for developers. MarkTechPost frames the release as frontier-class agentic coding and computer use at unchanged Opus pricing — a nuance worth noting: the price cut is relative to the pricier Fable 5 tier, not a reduction in what Opus itself has cost. Notebookcheck summarized it as more power for the same price, "but not for free."
Mashable reports Anthropic is billing Opus 5 as its safest model yet and has already made it the default on Claude Max, the company's highest-tier subscription, which starts at $100 a month. The Hindu businessline describes it as a more cost-efficient option aimed at everyday tasks.
The business backdrop matters too. Blockonomi notes the launch comes as Anthropic prepares for a 2026 public debut, making cost-per-capability a competitive talking point as much as a technical one.
Why it matters: the frontier of AI is increasingly being contested on price rather than raw capability, and a model that performs near the top tier for half the cost pushes advanced AI into far more everyday budgets — for individual developers, small companies, and anyone running AI agents at scale.