A8gent
GlossaryAgent concepts

What is claude Sonnet 5?

Also called: claude-sonnet-5, claude sonnet 5 pricing, sonnet 5 api price, anthropic sonnet 5

Claude Sonnet 5 is Anthropic's mid tier, with a 1,000,000-token context window and a standard API price of $3.00 per million input tokens and $15.00 per million output tokens.

Written by A8gent
Contents
  1. At a glance
  2. What Claude Sonnet 5 is
  3. What it costs, and the expiry you need to know about
  4. The arithmetic on a real agent
  5. When to move up or down
  6. What you should know
  7. FAQ
  8. Related terms

At a glance

Model ID
claude-sonnet-5
Standard input price
$3.00 per 1M tokens (rate table dated June 2026)
Standard output price
$15.00 per 1M tokens
Introductory input price
$2.00 per 1M tokens, through 31 August 2026
Introductory output price
$10.00 per 1M tokens, through 31 August 2026
Context window
1,000,000 tokens
Positioned as
Mid tier, the default for most production workloads

What Claude Sonnet 5 is

claude-sonnet-5 is the model ID for Anthropic's mid tier, and the one most production workloads should start on. It carries the same 1,000,000-token context window as claude-opus-5 at a lower price, which makes it the default rather than the compromise.

What it costs, and the expiry you need to know about

The standard rate is $3.00 per million input tokens and $15.00 per million output tokens.

Sonnet 5 also carries an introductory rate of $2.00 input and $10.00 output, documented as running through 31 August 2026. This is the single most important thing to know about this model ID before you build a budget on it: if you price a workload during the introductory window and assume it holds, your cost rises by 50 percent the moment it lapses. Model the standard rate, treat the discount as temporary, and diarise the date.

The arithmetic on a real agent

Take the same support agent shape: 4,000 tokens of system prompt and tool schema, 1,500 tokens of retrieved context, a 200-token message and a 400-token answer. At the standard rate that is 5,700 input tokens for $0.0171 plus 400 output tokens for $0.006, about 2.3 cents a turn. Ten thousand conversations of six turns is roughly $1,370 a month.

The same volume on claude-opus-5 is about $2,300. The same volume on claude-haiku-4-5 is about $560. Those gaps are why routing by task difficulty, rather than picking one model for everything, is usually the largest single saving available.

When to move up or down

Move down to Haiku 4.5 for classification, routing, extraction and short structured outputs, where the cheaper model usually scores the same on a real eval set. Move up to Opus 5 for long-horizon agent runs and hard reasoning where errors are expensive to catch downstream.

As always, the token bill is not the whole bill. An agent step costs one n8n execution or one Zapier task regardless of model, and that meter is a separate invoice.

What you should know

These figures come from Anthropic's published rate table dated June 2026, not from a live reading of the pricing page, which we could not reach when this entry was written. Check claude.com/pricing before committing a budget, and check specifically whether the introductory rate is still running.

Claude Sonnet 5 FAQ

How much does Claude Sonnet 5 cost?

The standard rate is $3.00 per million input tokens and $15.00 per million output tokens. An introductory rate of $2.00 and $10.00 is documented as running through 31 August 2026, so budget on the standard rate unless you have confirmed the discount is still live.

When does Claude Sonnet 5 introductory pricing end?

It is documented as running through 31 August 2026. After that the standard $3.00 and $15.00 rates apply, a 50 percent increase, so any workload priced during the introductory window needs re-checking.

What is the context window of Claude Sonnet 5?

1,000,000 tokens, the same as Claude Opus 5. Context window is not what separates the two tiers; capability and price are.

Should I use Sonnet 5 or Haiku 4.5?

Haiku 4.5 is a third of the price and usually scores the same on classification, routing and extraction. Start on Haiku for those, and reserve Sonnet for work where an eval set shows it actually wins.

Primary source

Anthropic pricing

Checked June 2026. Anything priced or rate-limited moves, so verify against the source before you budget.

Want this built rather than explained?

Knowing what claude sonnet 5 means is one thing; having it running in your business is another. We scope one workflow at a time, quote a fixed price in writing after a free consult, build it, and hand it over with documentation you own.