Anthropic Launches Claude Opus 5 as Near-Frontier Model at Half Fable 5 Price

Claude Opus 5 is a thoughtful and proactive model from Anthropic that comes close to the frontier intelligence of Claude Fable 5 at half the price.

Anthropic officially launched Claude Opus 5 on July 24, 2026. The model is positioned as a bridge between mid-tier and frontier capabilities in the company's lineup. It offers performance that comes close to Claude Fable 5 while maintaining a lower price point. The release includes immediate availability across multiple platforms. This strategy allows a wider range of users to access advanced AI tools for their projects. The pricing remains consistent with the previous Opus model to encourage adoption. Developers can begin using the model right away through the provided channels. The focus on thoughtful and proactive behavior makes it suitable for complex tasks that require sustained attention over long contexts. Enterprises evaluating AI options now have an additional choice that balances capability and cost without requiring a shift to the highest priced tier.

The launch expands options for teams that previously faced tradeoffs between performance and budget. Claude Opus 5 maintains the same token pricing as Opus 4.8 while delivering measurable gains on benchmarks that test reasoning depth. Users on Claude Max will see the model set as the default starting immediately. Claude Pro subscribers gain access to the strongest model currently offered under that plan. The combination of stable pricing and broad platform support reduces friction for organizations integrating the model into existing workflows. Cloud providers can now offer the model to their customers with the same day one rollout that Anthropic achieved on its own API.

What background led to the Claude Opus 5 release?

Anthropic has maintained a tiered approach to its model family that separates flagship frontier systems from more accessible variants. The introduction of Claude Opus 5 continues that pattern by placing a high capability model at the pricing level previously associated with the prior Opus generation. The decision reflects ongoing demand for models that support long running agents and professional workflows without the full expense of the top tier. Availability on AWS Bedrock adds enterprise grade features such as zero data retention that appeal to regulated industries. The model therefore serves both individual developers and large organizations that require secure deployment options from the outset.

Prior releases established the pricing baseline that Claude Opus 5 follows exactly. The company has kept input and output rates unchanged from Opus 4.8 while increasing the intelligence ceiling. This incremental improvement at constant cost provides a clear value proposition for customers already using the Opus line. The simultaneous rollout across four major access points further signals that Anthropic intends the model to reach production environments quickly. Background context from the models overview confirms that the pricing structure aligns with earlier Opus versions and remains half the rate of the Fable 5 flagship.

What are the technical specifics of Claude Opus 5?

Claude Opus 5 achieves performance within 0.5 percent of Fable 5 on CursorBench 3.2 when both models operate at maximum effort. The result holds while the new model incurs only half the cost per task. This benchmark measures coding and agentic capabilities that are central to professional use cases. The model retains the same input and output token rates as the previous Opus version. Those rates stand at five dollars per million input tokens and twenty five dollars per million output tokens. The pricing document lists identical figures for Opus 4.8 and doubles those amounts for Fable 5.

The technical profile supports extended agent sessions and complex multi step reasoning without forcing users into the higher cost bracket. Availability on AWS Bedrock includes compatibility with zero data retention configurations that protect sensitive data during inference. The model therefore meets requirements for both performance and compliance in a single package. These specifications position Claude Opus 5 as a practical choice for teams running sustained workloads where cost efficiency directly affects project viability.

How does the pricing of Claude Opus 5 compare to other models?

Pricing and positioning of selected Claude models

Model	Input Price per Million Tokens	Output Price per Million Tokens	Positioning
Claude Fable 5	$10	$50	Frontier intelligence
Claude Opus 5	$5	$25	Near-frontier at half cost
Claude Opus 4.8	$5	$25	Previous mid-tier version

The table above illustrates the direct cost advantage of Claude Opus 5 relative to Fable 5. Input and output rates remain unchanged from the prior Opus release. This continuity allows existing customers to upgrade performance without adjusting budgets. The half price differential versus Fable 5 applies uniformly across input and output tokens. Organizations that scale usage can therefore double their token volume for the same expenditure when switching from the flagship to the new Opus model.

Where is Claude Opus 5 available?

The model launched with day one support on four primary platforms. Access begins through the native Claude API for direct integration. Amazon Bedrock provides an enterprise channel with zero data retention options. Google Cloud and Microsoft Foundry complete the initial rollout. Each platform offers the same model weights and pricing structure. Customers can therefore select the hosting environment that best matches their existing infrastructure and compliance needs.

1. Claude API
2. Amazon Bedrock
3. Google Cloud
4. Microsoft Foundry

The ordered list above enumerates the launch platforms in the sequence provided by the official announcement. This multi platform strategy reduces vendor lock in and accelerates time to production. Teams already operating on AWS can adopt the model through Bedrock without additional infrastructure changes. The simultaneous availability across providers ensures that no single cloud service acts as a bottleneck for adoption.

What are the market and stakeholder implications?

The pricing and availability decisions affect how organizations allocate AI budgets across projects. Mid size teams gain access to near frontier performance that previously required flagship level spending. Large enterprises can deploy the model in long running agent scenarios while preserving zero data retention guarantees on Bedrock. The stable token rates relative to Opus 4.8 further simplify forecasting for finance and procurement teams. Overall the release broadens the addressable market for high capability models without increasing the cost structure.

Stakeholders in software development and professional services now have an additional tool for coding assistance and complex workflow automation. The performance level supports tasks that benefit from proactive model behavior while remaining within established pricing tiers. Cloud providers gain a new offering that can attract customers seeking advanced AI without premium costs. The combination of these factors suggests sustained interest in the Opus line as Anthropic continues to refine its model portfolio.

How have stakeholders reacted to the launch?

AWS highlighted the model as the most advanced Opus version yet and noted its compatibility with zero data retention. The statement emphasized benefits for coding, long running agents, and complex professional work. Anthropic described the model in its product announcement as thoughtful and proactive while stressing the half price advantage relative to Fable 5. These early reactions focus on the practical balance of capability, cost, and security features.

It’s a thoughtful and proactive model that comes close to the frontier intelligence of Claude Fable 5 at half the price.Anthropic

The pullquote captures the core positioning stated in the official announcement. AWS commentary adds the enterprise angle by referencing zero data retention and specific use cases. Together the statements indicate alignment between the model creator and a major distribution partner on the value delivered to end users.

What can users expect next from Anthropic?

The release establishes a template for future mid tier frontier models that maintain pricing continuity while closing the gap to flagship performance. Users can anticipate continued platform expansions as additional cloud providers integrate the model. The emphasis on agentic and professional workloads suggests ongoing optimization in those areas. Pricing stability across the Opus series provides predictability that supports long term planning for both individual developers and enterprise teams.

Subsequent updates will likely build on the benchmark results achieved on CursorBench 3.2. The multi platform availability model reduces barriers for new customers and supports experimentation at scale. Organizations monitoring the frontier models category now have a clearer path to deploy advanced capabilities within existing cost envelopes. The July 24 launch therefore sets expectations for how Anthropic will iterate on its model family going forward.