Home / Claude API pricing
Claude API pricing
What Anthropic charges for each Claude model through its API, per million tokens, and what a month of your own use would cost.
Checked October 13, 2026 against Anthropic’s pricing page
Your monthly Claude cost
Models
- Turn on JavaScript to see the costs. The price tables below work without it.
Current models
| Model | Input | Cached input | Cache write | Output | Batch in / out | Good to know |
|---|---|---|---|---|---|---|
| Claude Fable 5.1 | $10.00 | $0.25 | $12.50 | $50.00 | $5.00 / $25.00 | |
| Claude Opus 5.5 | $4.00 | $0.20 | $5.00 | $20.00 | $2.00 / $10.00 | |
| Claude Sonnet 5.5 | $2.00 | $0.10 | $2.50 | $10.00 | $1.00 / $5.00 | |
| Claude Haiku 5.5 | $0.10 | $0.01 | $0.13 | $0.50 | $0.05 / $0.25 | Over 100,000 tokens in a prompt: $0.50 in, $0.05 cached, $2.50 out |
US dollars per 1 million tokens, standard tier, from Anthropic’s pricing page, checked Oct 13, 2026.
Cache write is the 5-minute cache: 1.25 times the input price. The 1-hour cache costs 2 times the input price.
Older models still listed
| Model | Input | Cached input | Cache write | Output | Batch in / out | Good to know |
|---|---|---|---|---|---|---|
| Claude Opus 5 | $5.00 | $0.50 | $6.25 | $25.00 | $2.50 / $12.50 | |
| Claude Sonnet 5 | $2.00 | $0.20 | $2.50 | $10.00 | $1.00 / $5.00 | |
| Claude Opus 4.8 | $5.00 | $0.50 | $6.25 | $25.00 | $2.50 / $12.50 | |
| Claude Opus 4.7 | $5.00 | $0.50 | $6.25 | $25.00 | $2.50 / $12.50 | |
| Claude Opus 4.6 | $5.00 | $0.50 | $6.25 | $25.00 | $2.50 / $12.50 | |
| Claude Sonnet 4.6 | $3.00 | $0.30 | $3.75 | $15.00 | $1.50 / $7.50 | |
| Claude Sonnet 4.5 | $3.00 | $0.30 | $3.75 | $15.00 | $1.50 / $7.50 | |
| Claude Haiku 4.5 | $1.00 | $0.10 | $1.25 | $5.00 | $0.50 / $2.50 |
US dollars per 1 million tokens, standard tier, from Anthropic’s pricing page, checked Oct 13, 2026.
What changes the bill
Prompt caching
Reading a repeated part of the prompt from cache costs 10% of the input price on most models. Opus 5.5 and Sonnet 5.5 charge 5%, and Fable 5.1 charges 2.5%. Writing to the cache costs extra: 1.25 times the input price for 5 minutes, or 2 times for 1 hour. Anthropic notes that the 5-minute cache pays for itself after one read.
Batch
The Batch API halves input and output prices for requests that can wait. Anthropic says the caching discounts stack with the batch discount.
Long prompts on Haiku 5.5
Claude Haiku 5.5 is priced by prompt length. A request whose prompt is over 100,000 tokens pays $0.50 in and $2.50 out per million, five times the shorter-prompt price. Cache reads and writes count toward the 100,000, and each request is priced on its own. The other current models charge the same rate across their full 1-million-token context.
More tokens for the same text
Anthropic says Claude 4.7 and later models use a newer tokenizer that produces about 30% more tokens for the same text. When you compare per-token prices with older Claude models or other companies, the same job can use more Claude tokens than you expect. Count a sample with Anthropic’s token counting tool.
Other charges
- US-only processing (the
inference_geosetting) costs 1.1 times the normal price on Claude 4.6 and later models. - Fast mode, a research preview for Opus models, costs $8 in and $40 out per million on Opus 5.5. It does not work with batch.
- Web search costs $10 per 1,000 searches.
- Claude on Amazon Bedrock and Google Cloud is billed by those platforms at their own prices. Anthropic says their regional endpoints cost 10% more than global ones.
Not in the tables
Claude Mythos 5.1 and Mythos 5 are listed at the same prices as Fable 5.1 and Fable 5, but only for organizations in Anthropic’s verification programs. Claude Sonnet 5’s launch price of $2 in and $10 out became its standard price. The rise to $3 and $15 that had been planned for 1 September 2026 did not happen.