Claude Sonnet 4.6
BalancedLast verified: September 25, 2026· Anthropic Claude Sonnet 4.6 pricing
Previous-generation Sonnet, still served. Sonnet 5.5 is cheaper per token ($2 / $10). Adaptive thinking, 1M context.
- Input / 1M
- $3.00
- Output / 1M
- $15.00
- Context
- 1M tokens
- Provider
- Anthropic
Pricing verified September 25, 2026. Prices change frequently. Always confirm against the provider’s official pricing page before relying on these figures for budgeting. Official pricing →
When to choose Claude Sonnet 4.6
The workhorse, and for most products the model that should be carrying the bulk of production traffic. It pairs a 1M-token window with $3 in and $15 out, and caps the reply at 64K tokens rather than the 128K the Opus tier allows. Which routes genuinely need more than this is an evals question rather than an intuition one, and it is worth answering deliberately: the frontier tier is where budgets quietly go when nobody measures the boundary.
Claude Sonnet 4.6 sits close enough to Claude Haiku 4.5 on price that the two are usually decided together. That comparison is written once, on the models table, rather than argued twice from opposite sides.
What Claude Sonnet 4.6 costs in practice
The same 50M/10M month costs $300 here, against $163 on Gemini 2.5 Pro for an equally wide window and $100 on Haiku 4.5 for a fifth of the context. That spread is an argument for routing rather than for picking a favourite. Filling the 1M window costs $3 of input per request, so long prompts are affordable without being free, and retrieval still earns its place on routes that could technically skip it.
| Scenario | Input / mo | Output / mo | Est. cost / mo |
|---|---|---|---|
| Prototype | 2M | 0.5M | $14 |
| Growing product | 50M | 10M | $300 |
| At scale | 500M | 100M | $3,000 |
Plug in your own numbers with the cost calculator.
Where Claude Sonnet 4.6 fits in a routing setup
This is the row a routing setup usually gets built around, with traffic leaving it in both directions. Classification, extraction and routing itself go down to Haiku 4.5 at a third of the price. Genuinely hard reasoning goes up. Long-context reading, where that is the actual requirement rather than a habit, can go sideways to Gemini 2.5 Pro at $163 on the reference month against $300 here. The quality gate on each boundary is what makes the arrangement safe rather than merely cheap.
Cheaper alternatives to Claude Sonnet 4.6
The models immediately below Claude Sonnet 4.6 on blended cost, nearest first. A swap only pays if quality holds on the routes you move — see the LLM cost optimization playbook for how to prove that before you move them.
Frequently asked questions
How much does Claude Sonnet 4.6 cost?
Claude Sonnet 4.6 bills $3.00 per million input tokens and $15.00 per million output tokens — $6.00 per million blended at a three-to-one input:output split. A workload of 50M input and 10M output tokens per month works out at about $300.
What is Claude Sonnet 4.6's context window?
Claude Sonnet 4.6 has a 1M-token context window, of which up to 64K tokens can be the reply. One request that fills the whole window costs about $3.00 at the input rate, before the reply is billed.
Where does Claude Sonnet 4.6 sit on price?
At $6.00 per million blended — three input tokens for every output token — Claude Sonnet 4.6 is dearer than 14 of the 19 models tracked here and cheaper than 4. Output bills at 5x the input rate, and the nearest cheaper model on this table is GPT-4o at $4.38 blended.
Related concepts: context window, model routing, prompt caching, evals.