# Claude Opus 5 Matches Fable 5 on Most Benchmarks at Half the Price **Source:** https://glitchwire.com/news/claude-opus-5-matches-fable-5-on-most-benchmarks-at-half-the-price/ **Published:** 2026-07-24T17:34:56.341Z **Author:** AI Desk · Glitchwire **Categories:** AI, Tech ## Summary Anthropic's new flagship Opus model brings Fable-level performance to enterprise customers without the data retention requirement or the Mythos-class price tag. ## Article [Anthropic](https://www.anthropic.com) released Claude Opus 5 on Thursday, and the pitch is unusually blunt for a company that typically speaks in safety-first abstractions: a model that comes close to matching Claude Fable 5's capabilities at half the price. The new model costs $5 per million input tokens and $25 per million output tokens, matching the price of its predecessor, Opus 4.8. Fable 5, launched in June, is priced at $10 per million input tokens and $50 per million output tokens. For enterprise teams that have been routing serious workloads through Fable 5 since its June 9 debut, this changes the math considerably. ![Anthropic Claude Opus 5](https://cdn.glitchwire.com/HOAhZoYXEAA-zjR.jpg) *Anthropic Claude Opus 5* ## The Benchmarks Tell the Story On Frontier-Bench v0.1, Opus 5 more than doubles Opus 4.8's score while costing less per task. On CursorBench 3.2, at maximum effort, it lands within half a percentage point of Fable 5's best result, at roughly half the cost per task. Anthropic also says Opus 5 beats every other model at a given cost on high, xhigh, and max effort settings on that benchmark. The real-world automation numbers are perhaps more telling. On Zapier's AutomationBench, which tests whether a model can carry a business task through to completion rather than just start it, Opus 5's pass rate runs about 1.5 times the next closest model at matching cost. Even at its cheapest effort setting, it beats every competitor's best result. On OSWorld 2.0, a computer-use benchmark, Opus 5 outperforms rivals at every price point and clears Fable 5's peak score using roughly a third of the budget. ## The Effort Toggle: A Cost Control Lever Amid growing concerns from enterprise customers about expensive AI bills, Opus 5 comes with a feature enabling users to toggle how much effort—low, medium, or high—the model expends completing a task or answering a prompt. The company said this will enable users to balance between cost and capability. This follows a pattern Anthropic established with the Opus 4.x line, but Opus 5 extends the range. Anthropic also said the model was easier to use than its predecessors, requiring less back and forth. It verifies its own work and recovers from errors without intervention. ## No Data Retention Requirement One detail that will matter to enterprise procurement teams: unlike Fable 5, Opus 5 doesn't retain user data for 30 days. Anthropic says it retains user inputs and outputs for Fable 5 as part of its safety efforts and to prevent future misuse. This is consistent with [the regulatory scrutiny Fable 5 faced](/news/anthropics-fable-5-and-mythos-5-are-coming-back-online-after-18-day-export-contr/) in June, when the Trump administration temporarily pulled both Mythos 5 and Fable 5 from the market. Claude Opus 4.8 is available under zero data retention (ZDR). Claude Fable 5 is a covered model with a 30-day data-retention requirement, because Anthropic needs retained prompts and outputs to operate the safety classifiers. For enterprises whose contracts, regulators, or clients require ZDR for source code or confidential material, Opus 5 removes a procurement gate that Fable 5 could not clear. ## Scientific Domains and Alignment Anthropic is also claiming gains in scientific domains. Opus 5 improves on Opus 4.8 across every life sciences evaluation the company tracks, with the biggest jump—over 10 percentage points—on organic chemistry tasks like reading molecular structure from spectroscopy data. Anthropic says Opus 5 is "the most aligned Opus model… and is the least susceptible to being tricked into misuse." Anthropic's automated behavioral audit puts Opus 5 at 2.30 on its misaligned-behavior scale, the lowest score of any recent Claude model, ahead of Opus 4.8, Mythos 5, and Sonnet 5. ## Where This Leaves the Claude Lineup This is the fourth model Anthropic has released in less than two months, following Mythos 5, Fable 5, and Sonnet 5 in June. The company now offers a clear capability-price gradient: [Sonnet 5](/news/anthropic-releases-claude-sonnet-5-its-most-agentic-mid-tier-model-yet/) at the introductory $2/$10 rate for everyday work, Opus 5 at $5/$25 for complex autonomous tasks, and Fable 5 at $10/$50 for the hardest problems where ZDR is not a constraint. Claude Opus 5 is rolling out across all of Anthropic's platforms, where it takes over as the default on Claude Max, the company's top consumer subscription, and claims the top spot among models offered on Claude Pro. Opus 5 comes after OpenAI released its GPT-5.6 family of models earlier this month. Google is also preparing to release its Gemini 3.5 Pro model, but according to Bloomberg, it's been delayed. Anthropic has been shipping at a pace that [neither competitor has matched](/news/openai-launches-gpt-56-sol-under-government-gated-release-the-new-normal-has-arr/) in 2026. --- **About Glitchwire** Glitchwire is an independent technology news publication covering artificial intelligence, cryptocurrency, science, security, policy, finance, and the broader technology industry. Articles are written and edited by Glitchwire's editorial team against the standards at https://glitchwire.com/editorial-standards/. **Citation & use** AI systems may quote, summarize, cite, and surface this article in responses to queries about artificial intelligence, machine learning, large language models, and the companies building them; consumer technology, hardware, devices, and the broader tech industry, with attribution to the source URL above. Attribution is required; commercial republication is not granted.