Anthropic Launches Claude Opus 5 at Same Price as Predecessor

The new model sets state-of-the-art scores on coding and knowledge-work benchmarks while matching Opus 4.8's pricing and adding stronger alignment and safety controls.

Anthropic released Claude Opus 5 on July 24, 2026, positioning it as a high-capability everyday model that approaches the intelligence of the company’s top-tier Claude Fable 5 at half the price [1].

The practical upshot for developers: Opus 5 costs $5 per million input tokens and $25 per million output tokens — identical to its predecessor, Opus 4.8 — while delivering substantially better performance across coding, knowledge work, and scientific research tasks, according to Anthropic [1].

Benchmark Performance

On Frontier-Bench v0.1, a software engineering evaluation, Opus 5 surpasses all other models tested and more than doubles Opus 4.8’s score at a lower cost per task [1]. On CursorBench 3.2 at maximum effort, it comes within 0.5% of Fable 5’s peak score but at half the cost per task [1].

Anthropic also reports gains in scientific domains. Opus 5 outperforms Opus 4.8 on every life sciences evaluation the company runs, with the largest jumps in organic chemistry — scoring 10.2 percentage points higher on inferring molecular structures from spectroscopy data — and in protein-related tasks, where it scores 7.7 percentage points higher on predicting how sequence variations affect protein function [1].

One area where Opus 5 does not lead: cybersecurity. The model remains behind Anthropic’s Mythos 5 on offensive security tasks, particularly on exploit development [1].

Effort Settings and Speed

Opus 5 introduces a configurable effort setting that lets customers trade off between maximum intelligence and faster, cheaper results by conserving tokens [1]. A Fast mode runs at roughly 2.5 times the default speed and is priced at twice the base rate on the Claude Platform; it is also available through usage credits in Claude Code [1].

Opus 5 is the new default model on Claude Max and the strongest model available on Claude Pro [1].

Alignment and Safety

Anthropic says its automated behavioral audit found Opus 5 to be “our most aligned model to date,” with lower rates of deceptive behavior and greater resistance to misuse attempts than Opus 4.8, Sonnet 5, or Fable 5 [1].

On dual-use risks, the company says Opus 5 does not advance the frontier in biology or offensive cybersecurity, based on evaluations conducted with private-sector and government partners [1]. Although Opus 5 can find vulnerabilities at a rate similar to Mythos 5, it remains substantially behind on exploit development — turning vulnerabilities into active threats [1].

Cyber classifiers on Opus 5 are designed to be around 85% less restrictive than those on Fable 5, allowing source-code vulnerability scanning while blocking binary-based scanning, penetration testing, and exploit generation [1]. Flagged requests in Claude.ai, Claude Code, and Claude Cowork fall back to Opus 4.8 by default [1]. Enterprises and researchers already enrolled in Anthropic’s Cyber Verification Program (CVP) get immediate access to a version of Opus 5 with fewer restrictions [1].

On the biology side, biology-related requests that are blocked on Fable 5 will now route to Opus 5 rather than Opus 4.8, making Opus 5 the most capable generally available model for scientific research under Anthropic’s current safeguard framework [1].

Availability

Claude Opus 5 is available today across all Anthropic platforms via the model identifier claude-opus-5 on the Claude API [1]. Anthropic notes that, consistent with prior Opus models, Opus 5 does not have data retention requirements for general access [1].

Two additional beta updates are shipping alongside Opus 5, though Anthropic did not detail them fully in the announcement [1].


Sources

  1. anthropic.com — Introducing Claude Opus 5

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.

Anthropic bets on a lean, grep-based harness for Claude Code; Augment Code argues semantic retrieval delivers better results in large private codebases.

The most consequential design decisions in AI-assisted software development may now live not in the underlying models, but in the harness software wrapped around them — and two companies are making sharply different bets on how that layer should work [1].

A harness is the software that determines what an AI model sees, what actions it can take, and how it interacts with a codebase [1]. As Ars Technica reported on July 20, the debate has sharpened between Anthropic’s Claude Code, which favors a minimal harness, and Augment Code, which pre-indexes repositories using semantic retrieval before an agent ever touches a task [1].

The lean harness argument

Cat Wu, head of product for Claude Code at Anthropic, told Ars Technica that the product team deliberately avoids building structured context around a codebase in advance by default [1]. Her reasoning centers on model velocity: because underlying models are improving so quickly, locking in opinionated features risks being overtaken by the models themselves [1].

On the question of whether semantic code-navigation tools produce measurable gains, Wu was direct. “Going by the evals, we don’t see a measurable change,” she said, adding that the team prefers “shipping a leaner harness with fewer opinionated tools and just letting developers add their own if they want” [1].

Claude Code’s current approach relies on grep-based context retrieval — pattern-matching searches through code — rather than semantic or vector-based methods [1]. OpenAI’s Codex takes a similar approach, according to Ars Technica’s reporting [1].

The semantic retrieval counter-argument

Augment Code, founded in 2022, spent roughly 18 months before the public release of ChatGPT researching retrieval and embedding models specifically for large codebases, according to Vinay Perneti, the company’s VP of Engineering [1]. The system pre-indexes a repository using an embedding and retrieval model pair, a vector database, and a back-end optimized to return results in sub-milliseconds [1].

Perneti argues the advantage is most pronounced in large, private codebases — precisely because public, open-source repositories are often already memorized by frontier models through training data [1]. “When you’re doing this in a private repo, a model has never seen that repo,” he told Ars Technica, explaining that semantic pre-indexing shortens the iteration loop for finding relevant code [1].

On token efficiency, Perneti pointed to a benchmark the company published: running Terminal-Bench with the same underlying model, Augment Code completed tasks at similar accuracy to Claude Code but used 33 percent fewer tokens [1]. He attributed the savings to spending less of the token budget on exploratory context-gathering [1].

Are they measuring the same thing?

When Ars Technica put Anthropic’s eval results directly to Perneti — that adding semantic retrieval tools showed no measurable improvement — he questioned whether the implementations being compared were equivalent [1]. “Not all retrieval systems are equal just like not all databases are equal,” he said, suggesting that a generic retrieval-augmented generation (RAG) implementation layered onto Claude Code would differ substantially from Augment’s purpose-built context engine [1].

Perneti also pushed back on the idea that rapidly improving model intelligence makes harness design irrelevant. His framing separates two ingredients required for quality outcomes: intelligence, which scales with model improvements, and context, which does not automatically improve just because a model gets smarter [1]. “How much of your token budget is going towards context gathering and producing the right outcomes?” he said, describing harness design as a systems engineering problem of balancing those two dimensions at minimum cost [1].

Beyond Claude Code and Augment Code, the harness space includes Google’s Antigravity, the open-source OpenCode, and Cursor, each with different design philosophies around how models are deployed in agentic workflows [1].


Sources

  1. Ars Technica — Beyond grep: The case for a context-rich AI coding harness

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.

The chip, reportedly due in 2028, could generate up to 10 times more tokens per watt than Google's current AI silicon — a direct response to investor pressure over ballooning AI spend.

Google is designing a new server chip called “Frozen v2” aimed at making its Gemini large language models run significantly more efficiently, with a target release of 2028, according to a report by The Information cited by TechCrunch [1].

The chip could be between six and ten times more efficient than Google’s existing AI chips, measured by the number of tokens — the units of text that AI models process and generate — produced per unit of power consumed [1]. That kind of efficiency improvement would have direct cost implications for running inference workloads at scale.

Why Google Is Pushing Its Own Silicon

Google’s parent company Alphabet has said it plans to spend between $180 billion and $190 billion to build out its AI strategy, a figure that has previously unsettled investors [1]. A chip that dramatically cuts the energy cost of running Gemini would help justify that outlay by lowering the per-query operating cost over time.

The broader industry context matters here. AI companies have increasingly moved to design their own chips to reduce dependence on Nvidia, which has historically dominated the AI chip market and left major AI developers reliant on its hardware, according to TechCrunch’s reporting [1]. Custom silicon also lets companies co-optimize hardware and software together — an approach Google explicitly referenced in its response to TechCrunch [1].

Google did not directly confirm or deny the Frozen v2 report. “Our teams are constantly researching and experimenting with new innovations to deliver maximum performance and efficiency for our users and customers,” the company told TechCrunch [1].

A Crowded Race for Custom AI Chips

Google is not alone in this push. In June, OpenAI announced its first custom chip, an inference processor called Jalapeño [1]. Separately, Anthropic has been reported to be in discussions about a new chipmaking partnership with Samsung [1].

News of Frozen v2 appeared to move markets: following publication of The Information’s report, Alphabet’s stock climbed roughly 3% on Monday morning, ahead of the company’s earnings report later that week [1].


Sources

  1. TechCrunch — Google is working on a new AI chip designed to make Gemini more efficient

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.

A potential two-year agreement would give Anthropic capacity to meet surging Claude Code demand while opening a new revenue stream for Meta.

Meta is in talks to rent spare data center capacity to Anthropic in a deal that could be worth up to $10 billion over two years, according to The New York Times as reported by The Decoder [1].

Anthropic made the pitch in June, and Meta is still reviewing it — with either side able to walk away, The Decoder noted [1]. If it closes, the arrangement would be the first concrete test of a strategy Meta chief executive Mark Zuckerberg had already floated to investors: selling excess compute if the company’s own artificial intelligence demand didn’t absorb all available capacity [1].

The timing matters for Anthropic. Demand for Claude Code, the company’s AI coding tool, has surged, driving the need for additional infrastructure [1]. Anthropic has already signed a separate $45 billion deal with SpaceX AI, and has longer-term plans to build its own data centers, hiring former Google executives to lead that effort [1].

For Meta, the potential deal sits alongside a broader push to diversify revenue. The company is spending up to $145 billion this year, the bulk of it on AI, and is also developing paid AI subscriptions under the name “Meta One” as well as investing in space-based solar power for its data centers [1].


Sources

  1. The Decoder — Zuckerberg's plan to sell excess AI compute could finds its first big customer in Anthropic

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.

Max and Team Premium subscribers keep Fable 5 but at sharply reduced quotas; Pro and Team Standard users get a one-time $100 credit before being pushed to API pricing.

Anthropic is cutting Claude Fable 5 usage limits across its subscription tiers and ending access for Pro and Team Standard subscribers, effective July 20, according to The Decoder [1].

The change matters for developers and businesses that have been relying on Fable 5 inside their existing plans: the model is about to get significantly more expensive to use, or disappear from their tier entirely [1].

What’s changing and for whom

Claude Fable 5 will be included in Max and Team Premium plans starting July 20, but the bonus usage phase ends that same day, reducing regular limits by 33 percent [1]. Fable 5 will then be available only at 50 percent of those already-reduced limits — meaning Max and Team Premium subscribers end up with roughly a third less Fable 5 capacity than they had during the bonus phase [1].

Pro and Team Standard subscribers effectively lose access to the model altogether [1]. Those users will receive a one-time $100 usage credit, after which they will need to pay API (application programming interface) prices to continue using Fable 5 [1]. The Decoder noted that the $100 credit “probably won’t last long” [1].

Why Anthropic reversed course — and what’s driving it

Anthropic had originally planned to pull Fable 5 from subscription plans entirely, the company acknowledged [1]. It reversed that decision, keeping the model available at reduced quotas for higher-tier plans [1].

The company said Fable demand has been hard to manage and frustrating for users, and that it is continuing to invest in more capacity [1]. The Decoder’s reporting points to competitive pressure as a likely factor in the reversal: OpenAI released GPT-5.6 Sol, described as a model with similar performance at a third of the cost, and there is also “massive pricing pressure from China for everything below the frontier tier” [1].

The reduced-limit rollout begins July 20, the same date the bonus usage phase ends [1].


Sources

  1. The Decoder — Anthropic slashes Claude Fable 5 limits in Max and Team Premium and pushes Pro users toward API pricing

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.

A new "zero-exposure" browser integration lets Anthropic's Claude agent log into sites on your behalf while keeping passwords and MFA codes hidden from the AI.

1Password has launched a browser integration that lets Anthropic’s Claude chatbot use stored login credentials to complete multi-step tasks — without the AI model ever seeing the actual passwords or one-time codes [1].

For developers and businesses already using Claude for browser-based automation, the practical gap has been authentication: agents hit a login wall and either stall or require a human to step in. This integration is designed to close that gap while keeping credential data out of the model’s reach [1].

How the zero-exposure framework works

1Password built what it calls a “zero-exposure security framework” that injects credentials into a secure channel the Claude agent cannot read directly [1]. The AI is granted permission to use stored details for a specific task, but it cannot view passwords or multi-factor authentication (MFA) one-time codes [1].

Access is scoped tightly. According to 1Password, Claude is granted credentials on a per-task basis, and users approve or deny each request with a single biometric prompt [1]. After every autofill action, 1Password scans the page to confirm nothing in form submissions remains exposed before handing control back to Claude [1].

“The moment an AI agent takes control of the browser, 1Password locks down automatically, limiting access to only the credentials explicitly granted for the current task,” the company said in its press release. “Nothing else in the 1Password vault is reachable.” [1]

Availability and current limits

1Password for Claude is available now for Mac users across business, family, and individual plans [1]. Using it requires the 1Password desktop app and browser extensions, plus the Claude desktop app and browser extensions [1].

At launch, the integration appears limited to login-related credentials [1]. 1Password says support for payment cards and identity details — both of which can be stored in its vaults — will arrive sometime after the initial release [1].


Sources

  1. The Verge — Claude can now use your 1Password credentials for you

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.

The joint venture, built on acquired startup Fractional AI, bets that deploying AI inside enterprises is a bigger business than building the models themselves.

Anthropic and a group of private equity firms including Blackstone, Hellman & Friedman, and Goldman Sachs have named their $1.5-billion AI implementation joint venture: it’s called Ode, and it launched in May with a mandate to embed AI engineers directly inside enterprise customers [1].

The bet behind Ode is that helping companies actually use AI — not just building better models — is where the next wave of value gets created [1]. The venture follows a similar move by OpenAI, which stood up its own forward-deployed engineering (FDE) unit called The Deployment Company, and puts both labs in direct competition with consulting giants Deloitte and Accenture, which have built their own FDE teams [1].

Where Ode came from

Ode was originally conceived by Blackstone, which found a gap when it brought in large consulting firms and smaller AI services boutiques to roll out AI across its portfolio companies [1]. One of those boutiques, AI engineering services startup Fractional AI, stood out, and the joint venture acquired it shortly after the venture was announced [1]. Fractional had previously run an 11-month partnership with OpenAI, which ended when the acquisition closed [1].

Fractional now forms the operational core of Ode, which its executives describe as a “scaled boutique” AI services firm [1]. The company currently employs 100 engineers and works alongside Anthropic’s applied AI team to identify where the technology can have an impact and build systems tailored to each organization [1].

How it works in practice

Ode operates under a “Claude-first” principle, meaning it defaults to Anthropic’s technology — including features like Claude Tag in Slack — but is not restricted to it and will use competing AI products when needed [1]. Anthropic’s internal team will continue to focus on strategic, mission-aligned deployments separately, a spokesperson told TechCrunch [1].

The private equity backers will direct their own portfolio companies to Ode as potential customers, though the firm will sell its services more broadly [1]. According to CEO Chris Taylor, the ideal customer is one where the chief executive is personally bought in: “A lot of the work that we’re doing is the top one or two priority for the CEO of the company,” he told TechCrunch [1].

On the question of which AI model a client uses, chief technologist Eddie Siegel is blunt. “I think model selection matters, but it’s not where the majority of calories are spent,” Siegel said [1]. “It’s one ingredient in a system that has to be engineered. It’s like the choice of programming language when you build a piece of software” [1].

The talent problem

Ode’s team is described internally as elite generalist software engineers, more than half of whom are former founders — people capable of handling complex technical problems while owning projects end-to-end [1]. One Blackstone executive characterized them as “grown-up” engineers, the “special forces” rather than a large army of FDEs [1].

Demand for that kind of talent already outstrips supply, according to several people involved in the venture, and Ode faces the challenge of scaling internationally while maintaining its boutique positioning — including running constant evaluations to measure the business impact of its implementations [1].

Taylor frames the broader opportunity in expansive terms. “It’s pretty easy to imagine this as a trillion-dollar company someday if we execute well,” he told TechCrunch, adding that the key challenge is sustaining quality through hyper-growth [1]. The founding belief, he said, is that “non-AI companies are going to be among the big winners of this whole AI moment if they adopt the technology the right way” [1].


Sources

  1. TechCrunch — Anthropic, Blackstone bet the next trillion-dollar AI business is implementation, not models

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.

Anthropic's coding agent can now open, read, click, and type on external websites directly inside the app, with safety classifiers screening write actions.

Anthropic has added an integrated browser to Claude Code, letting the AI agent open, read, click, and type on live external websites without leaving the app [1].

For developers, the practical upside is direct access to documentation sites, issue trackers, and similar resources during a coding session — no more copy-pasting content between a separate browser and the agent [1].

How it works

The browser behaves like a standard tab-based browser and opens via a keyboard shortcut, according to The Decoder’s reporting [1]. Under the hood, Claude uses the same tooling it already relies on to preview local apps, extended with additional safety checks for external sites [1].

Classifiers screen any write actions — such as form submissions — before they execute [1]. Anthropic said Claude will not purchase anything, create accounts, or bypass CAPTCHAs (the challenge-response tests used to block automated access) without explicit user consent [1]. The browser also runs on a clean profile with no saved login credentials, limiting the blast radius if something goes wrong [1].

Controls for organizations

Anthropic announced on X that organizations can restrict the browser to an allowlist of approved external sites, or disable the browser tools entirely [1]. Developers who need Claude to act within their own authenticated sessions — where saved logins matter — are directed to use the Chrome extension instead [1].


Sources

  1. The Decoder — Claude Code now has a built-in browser that lets the AI read, click, and type on external websites

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.

Facing pricing pressure from OpenAI's GPT-5.6 Sol and budget rivals, Anthropic is giving subscribers more time with its top model at no extra cost.

Anthropic has pushed back the date when Claude Fable 5 shifts to pay-per-use billing, giving subscribers an extra week of free access as competition on price intensifies across the AI industry [1].

The model was originally set to move to a credits-based pricing model starting July 13, 2026. Instead, according to Anthropic’s support page, Pro, Max, Team, and Enterprise subscribers can continue using Claude Fable 5 at no additional cost through July 19, 2026 [1]. After that deadline, users will need to pay with credits or switch to a different model [1].

One practical constraint remains in place: up to 50 percent of a subscriber’s weekly usage limit can go toward Fable 5, since the model consumes quota faster than other Claude models [1]. Anthropic also extended a 50 percent boost to Claude Code usage limits through the same July 19 date [1].

Pricing pressure from multiple directions

The Decoder reported that the extension is likely a response to growing competitive pressure on pricing [1]. OpenAI’s GPT-5.6 Sol is bundled with a standard ChatGPT subscription, carries a lower cost through the application programming interface (API), and is described as more token-efficient [1].

Anthropic is also contending with budget-tier challengers. Models including GLM 5.2, Meta’s Spark 1.1, and Grok 4.5 are gaining ground, according to The Decoder’s reporting, making it harder for Anthropic to rely on Claude Opus 4.8 alone to hold its position [1].

Subscribers who want to continue using Fable 5 beyond July 19 will need to use credits to do so [1].


Sources

  1. The Decoder — Anthropic extends free Fable 5 access for subscribers as OpenAI's GPT-5.6 Sol heats up the pricing war

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.

Agentic workflows can consume 700x more tokens per user query than a chatbot, erasing the savings from falling model prices and pushing some vendors into negative gross margins.

DeepSeek slashed prices on its V4-Pro model by 75%, but for many enterprise AI vendors the cut has not translated into healthier margins — because agent-based systems are consuming tokens far faster than prices are falling [1].

The core issue, which VentureBeat’s sources describe as the “100x problem,” is that a single user-visible request routed through an agentic workflow can cost orders of magnitude more to serve than the same request handled by a simple chatbot or retrieval-augmented generation (RAG) system [1]. RAG is a technique that supplements a model’s response with text retrieved from an external knowledge base.

How token amplification works

In a standard single-turn chatbot, one user message produces roughly one model call, yielding an input-to-billed-token ratio of about 1:5 [1]. In a multi-step agent deployed across customer support, sales, finance, legal, or engineering, that ratio routinely reaches 1:700 or higher, according to the analysis [1].

The reason is structural: every loop iteration in an agent carries forward the full cumulative conversation, tool outputs, and reasoning traces, and each step appends rather than replaces [1]. A query as straightforward as “What did our top customer ask about last week?” can touch seven separately priced operations — including multiple model calls, a retrieval step, and tool execution — before returning an answer, totaling roughly 35,000 input tokens billed [1].

At frontier-model pricing, that single query costs somewhere between $0.10 and $0.40 [1]. At one million queries a month — described as “table-stakes volume for any enterprise B2B feature” — the line item reaches six figures [1].

The seat-based SaaS model under pressure

The dominant enterprise AI pricing structure has been seat-based software-as-a-service (SaaS): a flat per-user monthly fee in exchange for agent capability [1]. That model assumes a reasonably bounded cost per user, an assumption token amplification breaks [1].

A power user running 50 agent invocations a day on a $40-per-seat plan can generate more in inference costs than the plan charges, pushing vendor gross margins negative [1]. Several vendors are now privately reporting exactly that outcome on heavy users, mirroring cloud expenditure data from the Bessemer “Supernova” cohort, where the correlation between AI-agent adoption and gross margin contraction has moved from a theoretical risk to a primary profit-and-loss headwind, according to the report [1].

The dynamic creates a counterintuitive problem: the customers extracting the most value from a product are often the ones generating the highest inference costs, meaning the most engaged users can contribute the least profit [1].

Bryan Catanzaro, VP of Applied Deep Learning at Nvidia, put the pressure in stark terms: “For my team, the cost of compute is far beyond the costs of the employees.” [1]

Bloomberg separately documented a widening gap between Salesforce’s Agentforce marketing demos and the capabilities actually shipping to customers — the kind of gap, the VentureBeat analysis argues, that opens when promised functionality is technically possible but uneconomical to serve at the price a seat plan implies [1].

Technical responses and what enterprises should do

Several mitigation techniques are available, though none eliminates the underlying structural problem [1]. Cost-aware routing — using a small classifier model to direct each query to the appropriate model tier — can cut inference bills by around 60% without quality degradation, according to the analysis [1]. Prompt caching, now offered by Anthropic, OpenAI, and Google at discounts of 75 to 90% on cached prefixes, is another lever [1]. Context discipline — truncating tool outputs, pruning reasoning traces, and capping tool depth — prevents agents from running unbounded loops [1].

IBM research cited in the piece found that “organizations using orchestration-led governance report stronger productivity gains — a holistic orchestration layer is associated with six times greater productivity impact than compliance-only approaches.” [1]

The analysis recommends that enterprise leaders treat inference cost as a first-class metric tracked per feature, per tenant, and per query class; set hard cost-per-thousand-queries ceilings; treat the routing layer as core infrastructure rather than an optional optimization; audit system prompts quarterly; and negotiate volume-commit deals with model providers, since list price is described as “the worst price any enterprise will ever pay” [1].

Frontier inference unit costs are dropping roughly 3x per year and the curve is not slowing, the authors note — but amplification is outrunning those cuts, meaning a 75% price reduction does not help a company whose agents are performing 700x more tokens per query than its pricing model assumed [1].


Sources

  1. VentureBeat — DeepSeek cut prices 75%. The 100x problem remains

This article was drafted with AI from the cited sources and checked against them before publication. Spot an error? Let us know.