Now two months after the previous model update, Anthropic is sticking to its schedule. Claude Opus 5publishes its version.

According to the official statements made by Anthropic, the new generation Opus 5 performs very close to the frontier intelligence of Claude Fable 5, which the company kept for a long time due to cyber security concerns and is considered its most capable model.

The most striking part is that this high-level artificial intelligence performance is offered to end users and developers at exactly half the cost of the Fable 5 model.

Equipped with advanced data processing capabilities, optimized API processes and smart automatic routing systems, this new version aims to double efficiency while reducing costs for businesses that provide artificial intelligence integration.

Fable 5 Level Intelligence, Costs Half the Price

Anthropic classifies the artificial intelligence models it has developed into three different categories according to the needs of the market.

These; Haiku, the smallest and cheapest, Sonnet, which offers stable performance in the middle segment, and the Opus series, the largest and most powerful commercial model that can handle the most complex missions. With the update released today, the Opus model will be updated directly from version 4.8. opus 5transitions to level.

The biggest success of the new generation model lies in its cost efficiency. In the world of artificial intelligence, high process capacity often brings astronomical server costs.

While Anthropic defines the Opus 5 model as “thoughtful, proactive and efficient”, it states that this model offers a performance very close to the superior capabilities of the company’s flagship Claude Fable 5 model.

As a reminder, Fable 5 is known as Anthropic’s most advanced model, which was originally developed under the code name “Mythos” and was released with a delay by adding security protocols due to its enormous capabilities in cyber security and potential dangers.

Opus 5 offers an intelligence close to this enormous power at half the operating cost of Fable 5, greatly relieving the budgets of corporate companies and software developers.

Optimized for Daily Use: The New Default of Claude Max and Pro

It is of great importance that artificial intelligence models work quickly and at low cost so that they can be used in daily routine tasks. Anthropic designs the Claude Opus 5 model with an efficiency architecture that will serve exactly this purpose, that is, it can be used every day and without interruption.

The chip-friendly software infrastructure, which processes information in a much more optimized manner compared to other major language models, significantly reduces system response times.

As a result of these new hardware and software optimizations, Opus 5 provides enterprise-level Claude MaxIt becomes the new default model of the platform.

It is also preferred by personal professionals ClaudeProIt also takes its place as the most powerful and capable model under its subscription. Users can now perform all their powerful tasks, from code writing to complex information analysis, from long text summarization to proactive planning processes, with a much more fluid interface experience.

Two New Life-Saving API Features for Developers

In addition to innovations on the end-user side, Anthropic also introduces two very critical new functions for application developers using the Claude platform:

Mid-Conversation Tool Changes

Developers can now change the tools (functions and plug-ins) that artificial intelligence can use, right in the middle of an ongoing conversation on the Claude platform, without breaking or invalidating the prompt cache.

This innovation increases system performance, especially for software developers who develop dynamic and multi-stage artificial intelligence applications, while minimizing server costs and waiting times.

Automatic Fallbacks on API

Security filters and content classifiers can sometimes accidentally block sensitive but harmless requests sent by users.

Anthropic introduces the Automatic Fallback feature to solve this problem. Now, in Opus 5 or Fable 5 models, any API request that is caught or flagged by security filters (safety classifiers) is automatically directed to a safe alternative model by the system, instead of being blocked directly.

Thus, the system is not interrupted and API requests reach the best working model available without any problems, instead of being blocked all the time.

Claude Opus 5 Price and Availability

We are used to the fact that when there is an increase in performance in artificial intelligence models, prices generally move upwards. However, Anthropic does not change its price policy by following a user-friendly strategy.

The newly announced Claude Opus 5 model is now available simultaneously in all global markets and platforms supported by its prestige.

The pricing policy of the model remains exactly the same as the previous generation Opus 4.8. Accordingly, developers and corporate companies use Opus 5. $5 per 1 million input tokens, 25 dollars per 1 million output tokenscontinues to pay.

This new version, where performance is doubled, costs are maintained and process capability approaches the peak model, takes the competition in the artificial intelligence market to a much more exciting point.