# Claude Sonnet 5 explained: what Anthropic's cheaper AI agent means for you

> Anthropic's Claude Sonnet 5, out 30 June 2026, runs AI agents at near-Opus quality, far cheaper.

*In plain English: a faster, far cheaper way to run AI agents that plan and use tools for you — and why the fact it's bad at hacking is genuinely good news.*

By The SuggestedTech Team · SuggestedTech
Canonical: https://suggestedtech.com/news/claude-sonnet-5-explained

If you've heard people rave about "AI agents" and quietly wondered what the fuss is — or worse, seen the bill — this launch is worth five minutes of your time. On 30 June 2026 Anthropic released **Claude Sonnet 5**, which it calls its most agentic Sonnet model yet. The short version: it does the clever, autonomous stuff that until recently only the big, expensive models could do, but it costs a fraction as much to run. Here's what that actually means for you, in plain English.

## First, what is an AI agent — and why did running one get expensive?

A regular chatbot answers your question and stops. An **AI agent** keeps going: you give it a goal — "research these five suppliers and draft a comparison" — and it plans the steps, opens a browser, runs a terminal, checks its own work, and comes back with the finished job. Think of it as the difference between asking a colleague a question and handing them a task to complete while you do something else.

The catch is that all that back-and-forth costs money. Every step an agent takes sends text to the model and gets text back, and you pay per chunk of that text (a "token"). A ten-step task quietly costs ten times a one-shot answer. Through 2026 a lot of businesses got a nasty surprise when their agent bills landed — the capability was real, but so was the cost. That's the backdrop Sonnet 5 is walking into, and it's aimed squarely at it.

> **Info:** **In plain English:** an "agent" is an AI you give a *job* to, not just a question. It works in a loop — plan, act, check, repeat — and you're charged for every lap. So a model that's both good at the job *and* cheap per lap changes the maths completely.

## What Sonnet 5 changes: near-Opus quality at Sonnet prices

The headline is that Sonnet 5 gets you most of the way to Anthropic's flagship, Opus 4.8, for a lot less. On an agentic coding benchmark — a test of exactly the tool-using, multi-step work agents do — Sonnet 5 scored **63.2%**, against **69.2%** for Opus 4.8 and **58.1%** for the older Sonnet 4.6. On a separate knowledge-work benchmark, Anthropic says Sonnet 5 slightly *beat* Opus 4.8. In other words, you're not settling for a toy; you're getting something close to the top of the range for the price of the middle.

> "Opus 4.8 is still the model of choice for higher accuracy... but Sonnet 5 provides developers with lower-priced options... of much higher quality than what was previously available." Anthropic also notes the model's "effort level" is adjustable, so you can dial cost up or down against performance.
> — [Anthropic](https://www.anthropic.com/news/claude-sonnet-5), 2026-06-30

Here's how the money lines up. Prices are per **million tokens** (roughly 750,000 words), split into what you send in and what the model sends back out:

| Model | Price in / out | Notes |
| --- | --- | --- |
| Sonnet 5 (intro, to 31 Aug 2026) | **$2 / $10** | Launch price — the cheap window |
| Sonnet 5 (from 1 Sep 2026) | $3 / $15 | Standard price |
| Opus 4.8 | $5 / $25 | The pricier flagship |

So running the same agent on Sonnet 5 instead of Opus 4.8 can more than halve your cost, and during the introductory window it's cheaper still. Sonnet 5 is available on every plan and is now the **default model for Free and Pro** users, with Max, Team and Enterprise on board too — so most people are already using it without changing a thing.

## The two catches worth knowing before you get excited

Being honest with you: there are two things to keep an eye on. The first is a quiet one buried in the fine print — a **new tokenizer**. A tokenizer is just the way the model chops your text into billable pieces. Sonnet 5 chops differently, so the exact same document can count as anywhere from **1.0 to 1.35 times** as many tokens as before, depending on what's in it. That matters because your bill is per token, so a lower headline price doesn't automatically mean a lower invoice.

> Anthropic set the introductory pricing so that, for typical use, switching from Sonnet 4.6 to Sonnet 5 stays roughly cost-neutral despite the new tokenizer counting more tokens — though the real figure depends on your specific content.
> — [TechCrunch](https://techcrunch.com/2026/06/30/anthropic-launches-claude-sonnet-5-as-a-cheaper-way-to-run-agents/), 2026-06-30

So the practical advice is simple: if you're running real volume, don't take the sticker price on faith — run your own workload for a day and compare the actual token counts. The second catch is one of context rather than cost: nearly every performance number here comes from Anthropic's own launch post, and the early partner praise is from privileged access, not independent testing. The numbers are plausible and specific, but treat them as the maker's own homework until outside benchmarks confirm them. Anthropic even edited the post on launch day to fix a chart that had *undersold* Sonnet 5 — a reminder that these figures are freshly minted.

## Why "it's deliberately bad at hacking" is a good thing

This is the part I find genuinely reassuring. Anthropic made a deliberate choice **not** to train Sonnet 5 on cybersecurity offence — the skill of writing software that breaks into systems. As a result it's substantially weaker than Opus 4.8 at building exploits, and overall it behaves badly less often than Sonnet 4.6, especially when running as an autonomous agent. That's not the model failing at something; that's the model being *built* not to do something.

Why does that help you? Because an agent that can browse, click and run commands on its own is exactly the kind of tool you don't want to be a talented burglar. Choosing to leave the lock-picking skills out is a bit like selling a very capable delivery van that simply isn't wired to hotwire other cars — you lose nothing useful, and you sleep better. It's a pointed move, too: it landed the same month the US government pulled two of Anthropic's own models, Fable 5 and Mythos 5, over their cyber capabilities, so Anthropic is quietly making the safer choice a selling point. For an everyday user or a small team, a powerful agent that's boring to a criminal is precisely what you want.

## Key takeaways

- Claude Sonnet 5 launched on 30 June 2026 as Anthropic's most agentic Sonnet model, built to plan and use tools like browsers and terminals on its own.
- It costs an introductory $2 per million input tokens and $10 per million output tokens through 31 August 2026, then $3/$15 — well under Opus 4.8's $5/$25.
- On an agentic coding benchmark it scored 63.2% (Opus 4.8 got 69.2%, the older Sonnet 4.6 got 58.1%) and edged past Opus 4.8 on a knowledge-work benchmark, per Anthropic.
- A new tokenizer means the same text can count as 1.0–1.35x more tokens, so check your real bill; Anthropic set the intro price to keep switching roughly cost-neutral.
- Anthropic deliberately chose not to train Sonnet 5 on cybersecurity tasks, leaving it much weaker at building exploits than Opus 4.8 — a safety feature, not a bug.

## FAQ

### How much does Claude Sonnet 5 cost?
Through 31 August 2026 it's an introductory $2 per million input tokens and $10 per million output tokens, then $3/$15 from 1 September. For comparison, the pricier Opus 4.8 is $5/$25, so Sonnet 5 can more than halve your cost for similar agent work.

### Is Sonnet 5 as good as Opus 4.8?
Not quite, but it's close. Anthropic reports 63.2% versus Opus 4.8's 69.2% on an agentic coding benchmark, and says Sonnet 5 slightly beats Opus 4.8 on a knowledge-work benchmark. Opus stays the pick for maximum accuracy; Sonnet 5 is the value choice for running agents.

### What is the new tokenizer catch and will it raise my bill?
Sonnet 5 counts text differently, so the same content can map to 1.0–1.35x more tokens. Anthropic set the intro price so switching from Sonnet 4.6 stays roughly cost-neutral, but your real bill depends on your content — test your own workload before committing.

### Why can't Sonnet 5 hack, and is that a downside?
Anthropic deliberately didn't train it on cyber-offence, so it's much weaker than Opus 4.8 at building exploits and misbehaves less as an autonomous agent. For normal users that's a safety feature: a powerful agent that's useless to attackers, not a capability you'll miss.

### Do I need to do anything to use it?
Probably not. Sonnet 5 is available on all plans and is now the default model for Free and Pro users, with Max, Team and Enterprise supported. Developers can call it via the API using the model name claude-sonnet-5.

## Sources

- [Introducing Claude Sonnet 5](https://www.anthropic.com/news/claude-sonnet-5) — Anthropic, 2026-06-30
- [Anthropic launches Claude Sonnet 5 as a cheaper way to run agents](https://techcrunch.com/2026/06/30/anthropic-launches-claude-sonnet-5-as-a-cheaper-way-to-run-agents/) — TechCrunch, 2026-06-30
- [Anthropic Cuts AI Agent Costs With Claude Sonnet 5 Rollout](https://www.pymnts.com/news/artificial-intelligence/2026/anthropic-cuts-ai-agent-costs-with-claude-sonnet-5-rollout/) — PYMNTS, 2026-06-30