The Claude Sonnet 5 guide · 2026
How to use Claude Sonnet 5
Claude Sonnet 5 is only as good as the tool you run it through. We test and rank the best ways to put it to work — for coding, writing, building, chat and research.
Start here
New to Sonnet 5?
The basics
What is Claude Sonnet 5?
Claude Sonnet 5 is Anthropic's newest Claude Sonnet model, released June 30, 2026. It matches much of the intelligence of the top-tier Opus 4.8 at a fraction of the price, and it's especially strong at agentic work — planning, using tools like browsers and terminals, and running autonomously. It's the default model on Claude's Free and Pro plans and powers most of the best AI coding, writing and agent tools.
- Near-Opus intelligence at a fraction of the cost
- Built for agents — planning, tool use, and long autonomous runs
- The default model on Claude Free and Pro, and available via the API
Five ways to use it
Browse by use case
01 AI Coding Assistants Claude Sonnet 5 is widely regarded as the strongest model for real-world coding. These are the editors, agents, and CLIs that put it to work — ranked by how well they ship production code.
02 AI Writing Tools The best AI writing platforms now let you generate copy with Claude Sonnet 5 under the hood — combining Claude's nuance with templates, SEO workflows and brand voices. Here are the top picks.
03 AI App & Website Builders "Vibe coding" tools turn a prompt into a working app or site — and the best ones are built on Claude Sonnet 5. These platforms let anyone ship software fast.
04 Claude Chat Apps & Workspaces Want Claude Sonnet 5 at your fingertips everywhere — in your browser, on your desktop, or across your whole workspace? These AI assistants and workspaces put Claude (and other top models) one click away.
05 AI Research & SEO Tools From answer engines to full SEO suites, these research tools tap Claude Sonnet 5's reasoning to help you find, analyze and rank — faster.
From the timeline
What people are doing with it
Real posts from X, linked to the source — including the critical takes. We don't cherry-pick applause.
“Early testing for Claude Sonnet 5 showed: strong results across a range of coding scenarios, particularly for performance on CLI-style tasks; excellent prompt-cache utilization and competitive latency at lower effort levels.”
Rolling Sonnet 5 out in GitHub Copilot · Jun 30, 2026
“Sonnet 5 pairs frontier-level coding performance with a more affordable price point and outperforms Opus 4.8 on FrontierCode Extended.”
The team behind Devin, on adding Sonnet 5 · Jun 30, 2026
“In early tests, agents were more reliable, faster, and easier to trust with larger tasks than 4.6.”
Model marketplace, hands-on with launch traffic · Jun 30, 2026
“Claude Sonnet 5 is now available in Perplexity for Pro and Max subscribers. You can also select it as an orchestrator model in Computer.”
Answer engine, day-one availability · Jun 30, 2026
“Claude Sonnet 5 achieves 53 on the Artificial Analysis Intelligence Index, but without promotional pricing will cost more per task than Opus 4.8 — driven entirely by increased token usage.”
Independent benchmark lab — the honest cost caveat · Jun 30, 2026
“The same input text produces approximately 30% more tokens than on Claude Sonnet 4.6.”
Independent developer, from his launch-day write-up · Jun 30, 2026
“正直に言う。ベンチだけ見たら「地味」だと思った。使っていて思うのは、自律的にツールを使い続けるような多段タスクは確実に良くなってるし、ナレッジワーク系ではOpus 4.8を超えてくる。しかも比較的速くて安い”
"Honestly, from benchmarks alone I thought it was unremarkable. But in real use, multi-step tasks where it keeps using tools autonomously are clearly better, and on knowledge work it beats Opus 4.8 — while being comparatively fast and cheap."
Japanese AI-driven-development educator, after a day of testing · Jun 30, 2026
“Claude Sonnet5 vs Opus4.8 WEBサイト制作で比較。おススメの使い方: 1. Opus4.8で「0→1」を出す。 2. Sonnet5で細部の修正。 3. 最後にOpus4.8で最終チェック”
"Compared Sonnet 5 vs Opus 4.8 on website building. Recommended workflow: 1. Opus 4.8 for the zero-to-one. 2. Sonnet 5 for detail fixes. 3. Final check with Opus 4.8."
Japanese creator — a concrete mixed-model workflow · Jul 1, 2026
“Sonnet 5 の真価は Claude Code の Dynamic Workflows で出るとのこと - /model を Sonnet 5 にする - /effort を ultracode にする 試してみる”
"Apparently Sonnet 5's real value shows in Claude Code's Dynamic Workflows — set /model to Sonnet 5, set /effort to ultracode. Going to try it."
Japanese developer, on a Claude Code setup tip · Jul 1, 2026
From the journal