S&P 500100.00-1.70%NASDAQ112.50-0.85%Apple125.000.00%Microsoft137.50+0.85%Google150.00+1.70%Amazon162.50-1.70%Tesla175.00-0.85%Meta187.500.00%Bitcoin200.00+0.85%Ethereum212.50+1.70%EUR/USD225.00-1.70%Gold237.50-0.85%Oil250.000.00%
The Wiregazette
A man prepares to launch a model airplane on an outdoor runway, showcasing hobbyist passion.
Big Tech

Anthropic Launches Opus 5 at Half the Cost of Flagship Fable, Targeting Enterprise Budgets

4 min de lecture

Partager

Anthropic this week released Claude Opus 5, pricing it at half the cost of its top-tier Fable 5 model while matching or exceeding Fable on several benchmarks. The move signals a shift toward cost efficiency over raw capability as AI labs compete for enterprise customers.

Anthropic this week launched Opus 5, the latest version of its workhorse AI model, at half the price of its flagship Fable 5, according to the company. The pricing move marks a deliberate pivot toward affordability for enterprise users, with Opus 5 positioned as an everyday alternative to the more expensive and more heavily restricted Fable.

Opus 5 costs $5 per million input tokens and $25 per million output tokens, matching the pricing of its predecessor, Opus 4.8, and undercutting both Fable 5 and OpenAI's GPT-5.6, according to The Verge. Customers can also opt for a "Fast mode" at double the price, available in research preview.

The model launches just two months after Opus 4.8 became available on May 28, according to TechCrunch. It joins the 5-series lineup that already includes Mythos 5, Fable 5, and Sonnet 5, all released in June. Only the lightweight Haiku model remains without a 5-series upgrade, TechCrunch reported.

**Performance Gains Without a Radical Leap**

Opus 5 actually outperforms Fable 5 on a number of benchmarks cited in Anthropic's announcement, TechCrunch reported. Ars Technica noted that Opus 5 performs at roughly the same level or slightly ahead of Fable on coding tasks such as Frontier-Bench and DeepSWE, and beats Opus 4.8 and OpenAI's GPT-5.6-Sol across most tasks. The improvement is iterative rather than a breakthrough, Ars Technica wrote, with the primary pitch being near-Fable capability at approximately half the cost.

Anthropic emphasized in a post that Opus 5 is "much stronger at verifying its work and iterating carefully until it succeeds," offering examples of the model writing its own computer vision pipeline from an incomplete prompt, according to TechCrunch.

**Fewer Restrictions, Stronger Safeguards**

Opus 5 is free from several restrictions that have drawn scrutiny for Fable. Unlike Fable and Mythos, Opus 5 is not subject to the 30-day data retention policy that raised privacy concerns, according to both TechCrunch and Ars Technica. Anthropic expects the model's safety classifiers to engage 85% less often than for Fable 5, reflecting a lighter touch on the less capable model, TechCrunch reported.

However, meaningful cybersecurity safeguards remain. Opus 5 is blocked from scanning for vulnerabilities in software binaries, though it can search for vulnerabilities in source code, a task more likely to be used defensively, according to TechCrunch. Because Anthropic avoided giving Opus 5 cutting-edge training on cybersecurity, the model lags "substantially behind Mythos 5 on the exploitation of those vulnerabilities," Ars Technica reported.

Anthropic said Opus 5 is "the most aligned Opus model and is the least susceptible to being tricked into misuse," according to The Verge.

**New Automatic Fallbacks Feature**

Anthropic introduced a beta feature called Automatic Fallbacks, which automatically routes requests to a less powerful model when a user's prompt triggers the safety classifier, according to TechCrunch and The Verge. API users who enable the setting receive a functional response instead of an error message, reducing disruption.

**Enterprise Focus and Regulatory Context**

Anthropic is marketing Opus 5 especially to enterprise customers, calling it best for knowledge work and biology, and positioning it as the ideal everyday model, according to The Verge. Fable 5 remains the recommendation for the most ambitious tasks and long-horizon AI agent projects.

The launch comes weeks after Anthropic's regulatory negotiations with the Trump administration over Fable 5, which was briefly taken offline and reintroduced with stronger cybersecurity safeguards. When asked whether Opus 5 was reviewed by the administration before release, Anthropic spokesperson Danielle Ghiglieri told The Verge: "We continue to work with our government partners to conduct their own independent testing of our models. This includes Opus 5."

The pricing and capability mix positions Opus 5 as a volume play for Anthropic, offering enterprise customers a lower-cost alternative with fewer usage restrictions, while reserving the highest-end capability and cost for Fable.

Partager

À propos de Valentina Ortiz

Corporate Strategy Reporter. Covers how large companies allocate capital, enter new markets, and reposition for growth. She focuses on strategic pivots, partnerships, and operating decisions that affect shareholders.

Articles connexes

Close-up of a tablet displaying Google's search screen, emphasizing technology and internet browsing.