# Claude Sonnet 5

> Claude Sonnet 5 is the model most production workloads should default to. Anthropic describes it as its most agentic Sonnet — it plans, uses tools and runs autonomously at a level that until recently needed a larger and more expensive model. It has the same 1M-token context window and native prompt caching as Opus 5.

Canonical: https://routemux.com/anthropic/claude-sonnet-5
Provider: Anthropic

## Key facts

| Field | Value |
| --- | --- |
| Model ID | anthropic/claude-sonnet-5 |
| Official model ID | claude-sonnet-5 |
| Context window | 1M |
| Max output | 8K |
| Input modalities | text, image |
| Output modalities | text |
| Capabilities | Vision, Reasoning, Tool Calling, Coding, Cache, Strip Sampling Params |
| Upstream release | 2026-06-30 |
| Availability | available |

## Pricing

- RouteMux input: $0.56 per 1M tokens
- RouteMux output: $2.8 per 1M tokens
- Official input: $2 per 1M tokens
- Official output: $10 per 1M tokens
- Cache read: $0.056 per 1M tokens
- Cache write: $0.7 per 1M tokens
- Saving vs official: 72%
- Official price source: anthropic_x0.28, verified 2026-08-17

Billing is prepaid wallet, charged per successful request only.

## API access

Base URL: https://api.routemux.com

Supported protocols:

- anthropic_messages

## What is Claude Sonnet 5

Claude Sonnet 5 is the model most production workloads should default to. Anthropic describes it as its most agentic Sonnet — it plans, uses tools and runs autonomously at a level that until recently needed a larger and more expensive model. It has the same 1M-token context window and native prompt caching as Opus 5.

Through RouteMux it is $0.56 per 1M input tokens against Anthropic's $2, and $2.80 against $10 on output — 40% of Opus 5's rate for the same context window and the same Messages API.

The useful way to think about the Claude lineup here is that context is not the variable: Fable, Opus and Sonnet all carry 1M. What changes is how much reasoning you are paying for. Start on Sonnet and move up only when a task fails for reasoning reasons rather than context reasons.

## Key capabilities

- **Most agentic Sonnet** — Anthropic positions it for planning, tool use and autonomous runs — the workloads that used to require an Opus-class model.
- **Same 1M context as Opus** — Context is not what separates the Claude tiers; reasoning depth and price are.
- **Native prompt caching** — Reused system prompts and long shared context are billed at the cache rate.
- **40% of Opus 5's token price** — $0.56 vs $1.40 input, on the same API and the same context window.

## Other variants in this family

- Claude Sonnet 4.6 (`anthropic/claude-sonnet-4.6`)

## Sources

- [Anthropic — Models overview](https://docs.anthropic.com/en/docs/about-claude/models)
- [Anthropic — Pricing](https://www.anthropic.com/pricing)

_Content reviewed by Jerry Fan on 2026-08-17._