Guide8 min readPublished on 2026-07-24

Claude Opus 5 for coding: benchmarks, effort dial and what's new for agent builders

Claude Opus 5 for development: SWE-bench Pro at 79.2%, CursorBench close to Fable 5 at half the cost, the effort dial and the two API betas for agents. What changes in Claude Code.

In a nutshell

On coding Opus 5 makes a clear jump: SWE-bench Pro from Opus 4.8's 69.2% to 79.2%, and on CursorBench 3.2 at max effort it gets within half a point of Fable 5 while paying half the cost per task. It adds the effort dial to balance quality and spend, and two API betas — mid-conversation tool changes without losing the cache, and automatic fallbacks. For teams using Claude Code or building agents, it's the upgrade to make.

The coding jump, in numbers

Opus 5 is, by how the benchmarks tell it, the strongest flagship model on agentic coding. On SWE-bench Pro it scores 79.2% against Opus 4.8's 69.2%: ten points. For scale, Sonnet 5 on the same test sits at 63.2%, while the frontier models Fable 5 and Mythos 5 sit around 80%. On SWE-bench Verified the reported score is even higher.

A methodology note, because the numbers only count if they're honest: apart from SWE-bench Verified, these figures come from third-party reports in the first days after launch. The order of magnitude is solid, the individual percentages should be taken as indicative until the full official evals are out. If you need the non-technical picture, start from what Opus 5 is.

CursorBench and cost per task

The most interesting figure for developers isn't the quality peak, it's the quality/cost ratio. On CursorBench 3.2, at max effort, Opus 5 gets within about half a point of Fable 5 while paying half the cost per task. Anthropic also claims that at the high, xhigh and max effort levels, Opus 5 delivers more than any other model at the same cost.

Translated for a dev team: you can keep a flagship model on the whole flow — from reading the code to refactoring to debugging — without the electricity bill forcing you to downgrade on the heavy tasks. This is where the new model changes habits, not just benchmarks.

The effort dial for programmers

The effort dial is especially useful in development, where tasks differ a lot in difficulty. Generating a trivial unit test and designing a complex schema migration don't deserve the same reasoning budget.

With Opus 5 you raise the effort on architectural tasks and lower it on mechanical work. The result is a more predictable cost per session and, on hard tasks, more verification iterations before declaring done — Anthropic points precisely to a model "better at checking its own work and iterating until it succeeds." For anyone running agents autonomously, it's the difference between an agent that delivers and one that stops halfway.

Want to put Opus 5 to work on your code?

30 minutes to discuss your specific case.

Book a call

The two API betas that matter

Along with Opus 5, Anthropic shipped two beta features aimed at anyone building agents on the API.

The first: changing tools mid-conversation without invalidating the prompt cache. An agent can gain or lose a tool on the fly — for example unlocking database access only after a check — without paying for the whole context again. On long flows it's a real saving. The second: automatic fallbacks. If a request is blocked by the safety classifiers, it gets routed to another available model instead of failing. Fewer unhandled errors in production. These are the finishing touches that separate a prototype from an agent that holds up under real traffic. For the foundations, see the guide on Agent SDK and autonomous agents.

What changes in Claude Code

In Claude Code, Anthropic's terminal tool, Opus 5 becomes the recommended engine for serious work: more reliable on multi-file tasks, better at verifying and correcting before wrapping up, with effort you can raise when you need depth. Fast mode, faster at double the price, is handy when latency bothers you more than cost — typical of dense interactive sessions.

For a team already working in Claude Code, the upgrade doesn't require rewriting anything: change the model underneath and re-measure. If you're bringing Claude Code or agents into a business context, with control over permissions and security, look at Claude Code for business.

Is the upgrade worth it?

For anyone developing with Claude, yes, almost always. Same list price as Opus 4.8, clearly better coding performance, cost control with the effort dial and two betas that remove friction from agents. The one caveat: re-run your own tests and evals, because a new model changes the outputs and what was tuned before needs re-verifying.

At Maverick AI we build agents and integrations with Claude Code, Agent SDK and MCP, and bring them into production with the controls that are needed. If you want to put Opus 5 to work on your code or your agents, get in touch.

FT
Federico Thiella·Founder, Maverick AI

Works with European companies on Claude and Anthropic ecosystem adoption. Has led AI implementations in private equity, consulting, manufacturing and professional services.

LinkedIn

Want to put Opus 5 to work on your code?

We build agents and integrations with Claude Code, Agent SDK and MCP and bring them into production with the right controls. Get in touch.

Write to us

Frequently asked questions: Opus 5 for coding

Very: on SWE-bench Pro it scores 79.2% against Opus 4.8's 69.2%, and it's the strongest flagship model on agentic coding. On CursorBench 3.2 at max effort it gets within about half a point of the frontier models while paying half the cost per task. The numbers beyond SWE-bench Verified come from third-party reports and should be taken as indicative.
Yes. Opus 5 is available in Claude Code and is the recommended engine for serious development work, with the option to adjust the effort and use Fast mode in interactive sessions. You don't need to rewrite anything: change the model and re-measure.
Two features for anyone building agents: changing tools mid-conversation without invalidating the prompt cache, and automatic fallbacks that route requests blocked by the safety classifiers to another model instead of failing them. Both cut costs and interruptions in production.
In most cases yes: same price as Opus 4.8, better coding, cost control with the effort dial. The one caveat is re-running your own tests and evals, because a new model changes the outputs.
Fable 5 and Mythos 5 stay slightly ahead on the quality peaks, but Opus 5 gets close while paying roughly half the cost per task, and at the same cost it beats them. For a team's everyday development work, Opus 5 is almost always the choice with the best quality/price ratio.

Stay informed on AI for business

Get updates on Claude AI, business use cases and implementation strategies. No spam, just useful content.

Want to learn more?

Contact us to find out how we can help your company with tailored AI solutions.

Anthropic implementation partner in Italy. We work with companies in PE, pharma, fashion, manufacturing and consulting.

Related articles

News

Claude Opus 5: the flagship model now has a "dial" for cost — what changes for businesses

Anthropic has released Claude Opus 5: close to Fable 5 on many tasks at half the cost per task, with an "effort dial" that adjusts how hard the model reasons. Pricing, benchmarks and what it means for using AI in your company.

Technical

Claude Code: What It Is and How to Use It in Business

Claude Code is Anthropic's CLI tool that brings AI directly into the developer workflow. Learn what it is, how it works and how businesses use it for development, automation and code review.

Technology

Agent SDK: building autonomous AI agents with Claude

What is Anthropic's Agent SDK, how it works, how to design autonomous AI agents and what are the most promising enterprise applications.

News

Claude Sonnet 5: Opus-Level Performance at Sonnet Pricing — What Changes for Businesses

Anthropic has released Claude Sonnet 5: more agentic, close to Opus 4.8, and cheaper. Pricing, performance, and what it means for companies using AI.

Comparison

Opus 5 vs Opus 4.8: what really changes and whether it's worth switching

A comparison of Claude Opus 5 and Opus 4.8: same list price, better performance and the new effort dial. When it's worth switching and what to review in your setup.

Technical

Claude API: Integration Guide for Developers

Step-by-step Claude API integration guide: authentication, model selection (Haiku vs Sonnet vs Opus), Python SDK examples, rate limits and production-ready patterns for enterprise deployments.

Book an introductory call
Claude Opus 5 for coding: benchmarks and Claude Code (2026) | Maverick AI