No subscription, no minimum. Top up any amount, spend down at the per-1M rates below. Cache reads cost 10× less than fresh input — we surface every dimension.
| Model | Input ($/1M) | Output ($/1M) | Cache read ($/1M) | Cache write 5m | Context · max out | Capabilities |
|---|---|---|---|---|---|---|
|
Claude Opus 4.7
claude-opus-4-7
|
$15.00 | $75.00 | $1.50 | $18.75 | 200000 · 32000 | ×120/100 |
|
Claude Opus 4.1
claude-opus-4-1
|
$15.00 | $75.00 | $1.50 | $18.75 | 200000 · 32000 | ×120/100 |
|
Claude Sonnet 4.6
claude-sonnet-4-6
|
$3.00 | $15.00 | $0.30 | $3.75 | 1000000 · 128000 | ×120/100 |
|
Claude Sonnet 4.5
claude-sonnet-4-5
|
$3.00 | $15.00 | $0.30 | $3.75 | 1000000 · 128000 | ×120/100 |
|
Claude Haiku 4.5
claude-haiku-4-5
|
$1.00 | $5.00 | $0.10 | $1.25 | 1000000 · 64000 | ×120/100 |
|
Claude 3.7 Sonnet
claude-3-7-sonnet-latest
|
$3.00 | $15.00 | $0.30 | $3.75 | 200000 · 8192 | ×120/100 |
Every successful /v1/messages request emits an Anthropic message_delta with usage. We multiply the reported tokens by the rates above, apply our markup (already included in the per-1M column), and atomically debit your balance. Failed requests (4xx/5xx) are not charged.