claude-sonnet-4-5
- Input / 1M
- $3.00
- Output / 1M
- $15.00
- Cached input
- $0.30
- Context
- 200K
A 1,000-token prompt costs $0.00300 to send. Token counts for Anthropic models are estimated — approximated, because this family has no public tokenizer.
- Vision
- Tool calling
- Reasoning
- Max output 64K
Long-context rate
Past 200K tokens in a single request, claude-sonnet-4-5 switches to $6.00 input and $22.50 output per million — 2.0× the base input rate. Most calculators quote only the base figure.
Writing to the cache costs $3.75 per million at the 5-minute TTL, $6.00 at one hour. Reads are $0.30.
Pinned versions at the same rate: claude-sonnet-4-5-20250929
Priced closest to this
| Model | Provider | Input | Output | |
|---|---|---|---|---|
| claude-3-7-sonnet-20250219 | Anthropic | $3.00 | $15.00 | |
| claude-4-sonnet-20250514 | Anthropic | $3.00 | $15.00 | |
| claude-sonnet-4-20250514 | Anthropic | $3.00 | $15.00 | |
| claude-sonnet-4-6 | Anthropic | $3.00 | $15.00 | |
| gpt-3.5-turbo-16k | OpenAI | $3.00 | $4.00 | |
| mistral-large-2407 | Mistral | $3.00 | $9.00 |