Claude Sonnet 5: the new default, explained.
On June 30, 2026, Anthropic quietly swapped the engine behind most Claude conversations. The new Sonnet 5 gets close to flagship performance at a fraction of the cost — and if you're on the Free or Pro plan, you're probably already using it. Here's what changed, what it costs, and when the bigger models are still worth it.
01 What launched
On June 30, 2026, Anthropic released Claude Sonnet 5 — the new version of its midsize model — and made it the default model for Free and Pro users. If you use Claude.ai on a Free or Pro plan, your chats most likely switched to Sonnet 5 automatically. You didn't do anything wrong; nothing broke. The engine under the hood just got upgraded.
The pitch is simple: performance close to Opus 4.8 — Anthropic's heavyweight — at a fraction of the cost. The bigger story is what Sonnet 5 can do agentically: it plans multi-step work, uses tools like browsers and terminals, and runs autonomously at a level that a few months ago required the larger, more expensive models. Against February's Sonnet 4.6 it's a clear jump in reasoning, tool use, coding, and everyday knowledge work.
Most Claude users just got a near-flagship model as their everyday default — for free. The remaining question is when to reach past it.
02 What "default" means for you
Being the default matters more than the launch itself. Model pickers are like phone settings — most people never touch them. So the real-world effect of June 30 is that the median Claude conversation quietly became much more capable.
- Free plan: Sonnet 5 is now the model answering you. This is the biggest free-tier upgrade Claude has had.
- Pro plan: Sonnet 5 is your default, with the option to switch up to Opus 4.8 for the hardest problems.
- Max, Team, and Enterprise: Sonnet 5 is available alongside the rest of the lineup.
When should you switch up? Reach for Opus 4.8 when the task is genuinely hard — long multi-document analysis, intricate code architecture, high-stakes writing. And Fable 5, Anthropic's top-tier Mythos-class model, came back on July 1 after its June suspension — it's the premium option when you want the frontier and are willing to pay frontier prices.
03 Pricing and the 1M-token window
For API users, the model ID is claude-sonnet-5, with a 1-million-token context window — enough to load hundreds of pages into a single conversation.
- Introductory pricing: $2 per million input tokens / $10 per million output tokens, through August 31, 2026.
- Standard pricing after that: $3 input / $15 output per million tokens.
- For comparison: Fable 5 runs $10 input / $50 output per million — roughly 4–5× Sonnet 5's standard rate.
04 The Claude ladder, as it stands now
The lineup shifted twice in one week — Sonnet 5 arriving June 30, Fable 5 returning July 1 — so here's the current map, bottom to top:
- Haiku — fastest and cheapest; high-volume, lightweight tasks.
- Sonnet 5 — the new default. Strong enough for the large majority of real work, including agentic tasks.
- Opus 4.8 — the heavyweight for genuinely hard reasoning, coding, and writing.
- Fable 5 — the Mythos-class frontier model, back as of July 1 at premium pricing.
The practical habit: start on Sonnet 5, escalate only when it visibly falls short. Most people who default to the biggest model are paying (in money or rate limits) for capability their task never used. For the fuller cost picture, see what Claude actually costs — and for the Fable 5 saga, what happened to Fable 5.