Claude Haiku 3 – compact, near-instant.
tokens
tokens
Last Update
Overall Rating
| Type | Price | Unit | Context Type |
|---|---|---|---|
Input | $0.250000 | per mtokens | Standard |
Output | $1.250000 | per mtokens | Standard |
Cache Write (5 min) | $0.312500 | per mtokens | Standard Cache storage for 5 minutes |
Cache Write (1 hour) | $0.500000 | per mtokens | Standard Cache storage for 1 hour |
Cache Hit | $0.025000 | per mtokens | Standard Accessing cached content |
| Type | Price | Unit | Context Type | Savings |
|---|---|---|---|---|
| Input | $0.125000 | per mtokens | Standard | -50% |
| Output |
| Tier | RPM | TPM | IPM |
|---|---|---|---|
| tier_4 | 4,000 | 400,000 | 80,000 |
| tier_3 | 2,000 | 200,000 | 40,000 |
| tier_2 | 1,000 | 100,000 | 20,000 |
| tier_1 | 50 | 50,000 | 10,000 |
Batch processing allows for delayed, cost-efficient bulk operations
| Tier | RPM | TPM | Additional Limits |
|---|---|---|---|
| tier_4 | 4,000 | — | Queue:500,000 Requests Per Batch:100,000 Requests |
| tier_3 | 2,000 | — |
| per mtokens |
| Standard |
| -50% |
| tier_2 | 1,000 | — | Queue:200,000 Requests Per Batch:100,000 Requests |
| tier_1 | 50 | — | Queue:100,000 Requests Per Batch:100,000 Requests |
