Sonnet 5 — Chat with Anthropic’s New Claude Sonnet 5 Model in Your Browser
Sonnet 5 is a free Chrome extension that lets you chat with Claude Sonnet 5 — Anthropic’s newest AI model, released June 30, 2026 — right in your browser. It puts the new model one click away, with no keys and no configuration.
Open the extension and you get an instant chat: no API keys, no setup wizard, no juggling accounts. Ask anything and watch exactly what the new model can do, from quick answers to code and translations.
What Is Sonnet 5
Sonnet 5 is two things at once. The extension is a lightweight browser chat window, and the engine behind it is Claude Sonnet 5, the model Anthropic shipped on June 30, 2026. Understanding both makes it clear why the extension is worth installing.
A free browser chat, powered by Claude Sonnet 5
The extension runs entirely on Claude Sonnet 5, released by Anthropic as the successor to Sonnet 4.6. There is nothing to configure — click the icon, type a message, and the reply comes straight from the new model. You never touch a developer console or paste a token.
That matters because most people who hear about a fresh AI model have no easy way to try it. Sonnet 5 removes the friction: the same Anthropic Claude Sonnet 5 that developers reach through the API is available to you in a clean chat window, free to install from the Chrome Web Store.
The most agentic Sonnet yet
Claude Sonnet 5 is the most agentic Sonnet model to date. It can plan, reason, and use tools such as browsers and terminals, and it can coordinate subagents to run tasks autonomously — a level of capability that only months ago required larger, more expensive systems. It is, in Anthropic’s framing, an agentic AI model built to act, not just answer.
On launch day, the next-generation Sonnet became the default model across Claude’s consumer plans and is available to Max, Team, and Enterprise users. When you chat through the extension, you are talking to the same Sonnet-class model that now powers those experiences.
Why Claude Sonnet 5 Is a Big Deal
The headline is simple: Claude Sonnet 5 approaches the quality of Anthropic’s flagship Opus 4.8 while costing far less to run. That combination of near-flagship performance and lower price is what makes this release stand out.
Near-Opus 4.8 quality
On agentic coding benchmarks, Claude Sonnet 5 scores 63.2%, versus 69.2% for Opus 4.8 and 58.1% for the previous Sonnet 4.6. The gap to the flagship narrows sharply, and on some knowledge-work benchmarks the new Sonnet slightly surpasses Opus 4.8.
The generational jump from Sonnet 4.6 to Sonnet 5 is the real story. A five-point climb on hard agentic coding tasks is the kind of gain that changes what a mid-tier model can realistically handle, from multi-step research to writing and fixing code.
| Model | Agentic coding | Positioning |
|---|---|---|
| Opus 4.8 | 69.2% | Anthropic flagship |
| Claude Sonnet 5 | 63.2% | New mid-tier, near-flagship |
| Sonnet 4.6 | 58.1% | Previous Sonnet |
Self-correcting and safer by design
Claude Sonnet 5 reviews its own answer and fixes errors before you ever see them. This self-correction means fewer obvious mistakes slip through, which is exactly what you want from a model you are chatting with casually rather than double-checking line by line.
Safety improved too. Anthropic reports better refusal behavior and stronger prompt-injection resistance, and both new Sonnet models scored 0.0% on the Firefox 147 exploit benchmark — work done with Mozilla and patched in Firefox 148.
Claude Sonnet 5 is our most capable Sonnet model yet, built to run agents reliably and safely at a lower cost.
Anthropic
Pricing and How It Compares
Cost is where Claude Sonnet 5 makes its clearest argument. Anthropic priced it deliberately below both its own flagship and the strongest models from OpenAI and Google, and the same intelligence is available across the company’s developer tools.
Lower cost than rivals
Introductory API pricing is $2 per million input tokens and $10 per million output tokens through August 31, 2026, then $3 / $15 after that, as listed in the Claude Platform pricing docs. That sits well below Opus 4.8 at $5 / $25, and undercuts OpenAI’s GPT-5.5 and Google’s Gemini 3.1 Pro. Anthropic positions the model as a direct, lower-cost alternative to GPT-5.6 Sol and Gemini 3.5 Flash for everyday agentic tasks.
The table below shows how the input/output pricing stacks up against Anthropic’s own flagship.
| Model | Input (per 1M) | Output (per 1M) |
|---|---|---|
| Claude Sonnet 5 (intro) | $2 | $10 |
| Claude Sonnet 5 (from Sep 1, 2026) | $3 | $15 |
| Opus 4.8 | $5 | $25 |
Independent coverage framed the launch the same way, with TechCrunch describing it as a cheaper way to run agents.
The same model developers use
The extension gives you the same Sonnet-class intelligence that developers reach through Claude Code and the Claude Platform and API. There is no developer setup on your side — the model is identical, only the interface is simpler.
That is a genuinely useful shortcut. If you are curious whether claude-sonnet-5 fits your workflow before committing to an API integration, the extension lets you pressure-test the model in plain chat first, then move to Claude Code or the API when you are ready.
What You Can Do in the Extension
The chat window is deliberately plain, and that is the point — you spend your attention on the conversation, not the tool. In day-to-day use it covers most of what people reach an AI assistant for:
- Chat instantly and ask questions, explanations, or brainstorming prompts
- Get quick summaries of long or dense material
- Write and rewrite emails, posts, and code snippets in seconds
- Paste any text and have Sonnet 5 explain, translate, or condense it
Because it lives in your browser, the model is always a click away while you read, write, or research — no switching tabs to a separate app.
Who It’s For
Anyone who wants to try Claude Sonnet 5. If you have read the launch news and want to see the new model for yourself, the extension is the fastest path — free, instant, and no account gymnastics.
Writers and students who want a fast assistant. Drafting, summarizing, and rewriting are exactly the everyday tasks the model handles well, and the self-correction helps keep the output clean.
Developers exploring an agentic Sonnet-class model. Before wiring up the API, you can feel out how the model reasons and where its limits are, using the same intelligence available in Claude Code.
Professionals who need quick answers while they work. Quick explanations, translations, and rewrites are one click away without breaking your flow.
How to Use
Getting started takes under a minute:
- Install Sonnet 5 from the Chrome Web Store.
- Click the icon in your toolbar to open the chat.
- Type your message or paste the text you want help with.
- Get an instant answer from Claude Sonnet 5.
That is the whole workflow — there is no onboarding, no key to generate, and nothing to pay for.
Frequently Asked Questions
- What is Sonnet 5?
Sonnet 5 is a free Chrome extension that lets you chat with Anthropic’s Claude Sonnet 5 model right in your browser, with no setup or API keys.
- What is Claude Sonnet 5?
Claude Sonnet 5 is Anthropic’s most agentic Sonnet model, released June 30, 2026 as the successor to Sonnet 4.6, offering near-Opus 4.8 quality at a lower cost.
- When was Claude Sonnet 5 released?
June 30, 2026. On launch day it became the default model across Claude’s consumer plans.
- How much does Claude Sonnet 5 cost?
API pricing is $2 / $10 per million input/output tokens through August 31, 2026, then $3 / $15. The extension itself is free to install.
- Is Sonnet 5 better than Opus 4.8?
Claude Sonnet 5 approaches Opus 4.8 — 63.2% versus 69.2% on agentic coding — and slightly surpasses it on some knowledge-work benchmarks, at a much lower price.
- Where else is Claude Sonnet 5 available?
Besides this extension, the same Sonnet-class model is available in Claude Code and the Claude Platform and API, and it powers Claude’s Free, Pro, Max, Team, and Enterprise plans.