Claude Opus 4.8 vs Claude Sonnet 5.5
Claude Sonnet 5.5 is one of Anthropic's newest models and Claude Opus 4.8 an earlier one, so this is also what moving up costs or saves. Claude Sonnet 5.5 is the cheaper of the 2: for the same tokens, 3 read for each 1 written, Claude Opus 4.8 costs 2.5 times as much. Below are both price lists, what 8 everyday jobs cost a month on each, and how many tokens the same text takes on each, which can change the answer.
- For the same tokens
- Claude Sonnet 5.5 is cheaper; Claude Opus 4.8 costs 2.5 times as much
- Cheaper to read
- Claude Sonnet 5.5, $2 against $5
- Cheaper to write
- Claude Sonnet 5.5, $10 against $25
- Scores higher
- Claude Sonnet 5.5, 164 against 158
- Larger window
- The same, 1M tokens each
Price per million tokens, side by side
In US dollars, as each maker prints them. The better figure in each row is green.
| Claude Opus 4.8Anthropic · earlier model | Claude Sonnet 5.5Anthropic | |
|---|---|---|
| InputTokens the model reads | $5 | $2 |
| Cached inputRead again from the cache | $0.50 | $0.10 |
| Cache writePutting text in the cache | $6.25 | $2.50 |
| OutputTokens it writes, thinking included | $25 | $10 |
| Blend 3:13 tokens read for each 1 written | $10 | $4 |
| Batch inputAnswers collected later | $2.50 | $1 |
| Batch output | $12.50 | $5 |
| Context windowTokens it holds at once | 1M | 1M |
| Capability scoreEpoch Capabilities Index; GPT-5 is 150 | 158 | 164 |
| Longest answerTokens it writes in 1 reply | 128K | 128K |
Price changes
- Claude Sonnet 5.5: Cached input went from $0.20 to $0.10 per million tokens. The price of a cache write did not change. What changed
What 8 everyday jobs cost a month on each
Each job at its usual number of runs a day, for 30 days, at the standard price. The cheaper of the 2 is green; open any job in the calculator to change it.
| Job | Claude Opus 4.8 | Claude Sonnet 5.5 | Difference |
|---|---|---|---|
| A chat question300 in, 500 out, 100 a day | $42.001.4 cents a run | $16.800.56 cents a run | 2.5 times |
| A hard question, with thinking300 in, 1,400 out, 100 a day | $1103.6 cents a run | $43.801.5 cents a run | 2.5 times |
| A long report, summarised60,000 in, 2,000 out, 20 a day | $21035 cents a run | $84.0014 cents a run | 2.5 times |
| A 40-turn agent task2,800,000 in, 40,000 out, 10 a day | $1,098$3.66 a run | $364$1.21 a run | 3 times |
| A support bot's reply4,000 in, 300 out, 2,000 a day | $1,1101.8 cents a run | $4320.72 cents a run | 2.6 times |
| Sort a message into a category250 in, 5 out, 50,000 a day | $2,0630.14 cents a run | $8250.055 cents a run | 2.5 times |
| Read an invoice into fields1,500 in, 250 out, 1,000 a day | $4131.4 cents a run | $1650.55 cents a run | 2.5 times |
| Review a pull request25,000 in, 4,500 out, 50 a day | $35624 cents a run | $1439.5 cents a run | 2.5 times |
Your own job on both
Start from a job, then set its size, how often it runs and whether the answers can wait. Both bills follow every change.
A short question and a few paragraphs back.
about 230 words
Instructions or documents the model has read before cost far less the second time.
about 380 words
A month of this job
- Claude Opus 4.8$42.00
1.4 cents a run
- Claude Sonnet 5.5$16.80
0.56 cents a run
Claude Sonnet 5.5 saves $25.20 a month; Claude Opus 4.8 costs 2.5 times as much.
This job on every modelWhy the price per token is only the start
Every maker prints 2 main prices: one for the tokens a model reads (input) and one for the tokens it writes (output). On most current models writing costs 3 to 5 times as much as reading, so which model is cheaper can depend on the job. A job that reads a long document and writes a line pays mostly the input price; a chat that writes paragraphs pays mostly the output price. That is why each comparison prices 8 jobs and not one figure.
The same text is not the same number of tokens on every model. Each maker cuts text its own way, and Anthropic says its models from Claude 4.7 on produce about 30 percent more tokens for the same text than earlier ones. A model with the lower price per token can still send the larger bill, so each comparison counts one of our lessons the way both makers count it.
Discounts change the order too. Work that can wait costs half on a batch at the makers that offer one, and text read again from a cache costs a quarter of the input price or less on every current model here. The price table on each comparison shows who offers what.
Thinking is billed as output, and models differ in how much they think and how long they answer. A price list cannot show that. Before you switch, run the same 10 tasks on both models and compare the bills.
Questions
- Which is cheaper, Claude Opus 4.8 or Claude Sonnet 5.5?
- Claude Sonnet 5.5 is cheaper: $2 per million tokens read and $10 per million written, against $5 and $25 on Claude Opus 4.8. On a mix of 3 tokens read for each 1 written, Claude Opus 4.8 costs 2.5 times as much.
- Which is better, Claude Opus 4.8 or Claude Sonnet 5.5?
- On the Epoch Capabilities Index, an independent score from Epoch AI that combines over 50 benchmarks, Claude Sonnet 5.5 scores 164 and Claude Opus 4.8 158, above the top of the range of doubt around Claude Opus 4.8's score (GPT-5 is 150). One score is an average; test both on your own work before you switch.
- What does a support bot cost a month on Claude Opus 4.8 and Claude Sonnet 5.5?
- At 2,000 replies a day, each reading 4,000 tokens (50 percent of them from cache) and writing 300, it costs $1,110 a month on Claude Opus 4.8 and $432 on Claude Sonnet 5.5: Claude Opus 4.8 costs 2.6 times as much for this job.
- What does an AI agent task cost on Claude Opus 4.8 and Claude Sonnet 5.5?
- One 40-turn task that reads 2,800,000 tokens in all, 90 percent of them from cache, and writes 40,000, costs $3.66 on Claude Opus 4.8 and $1.21 on Claude Sonnet 5.5. At 10 tasks a day that is $1,098 and $364 a month.
- Which has the larger context window, Claude Opus 4.8 or Claude Sonnet 5.5?
- Neither: both hold 1M tokens at once, about 1,500 pages of English text.
- Does the same text take the same number of tokens on Claude Opus 4.8 and Claude Sonnet 5.5?
- Yes, both are counted the same way: our lesson "How AI writes", 1,700 words long, is about 2,100 to 3,600 tokens on each.
- Is it cheaper to move from Claude Opus 4.8 to Claude Sonnet 5.5?
- Yes, per token: Claude Sonnet 5.5 costs $2 per million tokens read and $10 written, against $5 and $25 on Claude Opus 4.8, which costs 2.5 times as much for the same tokens. Run your own tasks on it before you switch, since answers, length and thinking can differ.
- Can I pay less on Claude Opus 4.8 and Claude Sonnet 5.5 for work that can wait?
- Yes, on both. With a batch, where you collect the answers later instead of at once, Claude Opus 4.8 costs $2.50 per million tokens read and $12.50 written, and Claude Sonnet 5.5 $1 and $5.
Sources