AI cost comparison
Opus vs Sonnet cost calculator
Opus 5.5 ($4 in, $20 out per million tokens) costs exactly twice Sonnet 5.5 ($2 and $10) on every line of the price list. Whether it costs twice as much per task depends on how many tokens each model needs and how often Sonnet's first attempt has to be redone.
Your inputs
Result
Opus 5.5 $144 vs Sonnet 5.5 $72.24 a month
Opus costs 2.00 times as much here. It becomes the cheaper model only if it needs under 50% of Sonnet's tokens per task.
- Sonnet 5.5 per task
- $0.07
- Opus 5.5 per task
- $0.14
- Sonnet 5.5 per month
- $72.24
- Opus 5.5 per month
- $144
- Extra cost of Opus per month
- $72.24
- Break-even Opus token share
- 50%Opus is cheaper below this
Worked examples
Each row is the calculator run on the inputs shown, with every other input at its default.
| Scenario | Opus token share | Sonnet redo rate | Sonnet per month | Opus per month |
|---|---|---|---|---|
| Same tokens, no redos | 100% | 0% | $72.24 | $144 |
| Opus needs 30% fewer tokens | 70% | 0% | $72.24 | $101 |
| Opus needs half the tokens | 50% | 0% | $72.24 | $72.24 |
| A quarter of Sonnet tasks redone | 100% | 25% | $90.30 | $144 |
| Hard tasks: Opus 60%, Sonnet redo 40% | 60% | 40% | $101 | $86.69 |
How it is calculated
Opus 5.5 costs twice Sonnet 5.5 for input, output, cache writes and cache reads, so for the same tokens it always costs twice as much. The calculator lets the token counts differ.
With no redos the break-even share is 50%: Opus is cheaper only if it finishes a task with less than half of Sonnet's tokens. Every Sonnet task that needs a second attempt raises that line. If a quarter of Sonnet tasks are redone, Opus breaks even at about 62% of Sonnet's tokens.
The token share and redo rate are yours to measure: run a sample of real tasks on both models and compare the token counts in your usage logs. Claude 4.7 and later share one tokenizer, so the same text makes the same number of tokens on Opus 5.5 and Sonnet 5.5.
Questions
How much more expensive is Opus than Sonnet?
Opus 5.5 costs exactly twice Sonnet 5.5 per token: $4 against $2 per million input tokens, $20 against $10 output, and twice for cache reads and writes.
When is Opus cheaper than Sonnet?
When it uses less than half of Sonnet's tokens for the same task, or less than a higher share if Sonnet often needs a second attempt. The break-even line above moves with your redo rate.
Should I use Opus or Sonnet for Claude Code?
Many teams use Sonnet by default and switch to Opus for hard, long tasks where it finishes in fewer turns. The calculator shows what that switch costs per month.
How does Fable 5.1 compare?
Fable 5.1 costs $10 input and $50 output per million tokens, 2.5 times Opus 5.5, with cache reads at $0.25. It appears in the chart above at Sonnet's token counts.
Sources
- Anthropic: Claude API pricing, checked 2026-10-29
Prices and plans change. This page was last checked on 2026-10-29. The data is open: llm-prices.json (CC BY 4.0).