Haiku 5.5 API Pricing
Haiku 5.5 has tracked API rates of $0.1 per 1M input tokens and $0.5 per 1M output tokens. Cached input is billed at $0.01 per 1M tokens. Rates are in USD from the AI IQ dataset; see the Haiku 5.5 model profile for capability scores.
Price List
| Rate | Price (USD) | Notes |
|---|---|---|
| Input | $0.1 | per 1M tokens |
| Output | $0.5 | per 1M tokens |
| Cached input (read) | $0.01 | per 1M tokens |
| Cache write | $0.125 | per 1M tokens |
| Cache write (1-hour) | $0.2 | per 1M tokens |
| Blended input + output | $0.6 | 1M in + 1M out at list price |
| Context window | 1M tokens |
Pricing References and Billing Caveats
These are the tracked rates, not a live quote. Hosted-provider choice, region, long-context tiers, caching, batch discounts, and temporary promotions can change the actual bill. Open-weight availability does not mean hosted inference is free. Workload examples assume the listed standard input and output rates, with no cache hits or discounts.
Anthropic pricing catalog — check current offerings and rate tiers. This catalog may not list historical models or third-party hosted endpoints.
Publisher-reported pricing notes
- Higher pricing applies to prompts over 100,000 tokens; listed token prices use the lower prompt-length tier.
- Effort settings are configurations of one model, not separate commercial models.
Official release and model references
These references document the model and its release; not every reference is a pricing table. Confirm the applicable endpoint and billing tier with your inference provider before purchase.
What a Workload Costs
Computed straight from the list prices above — token counts are illustrative workload sizes, not measurements of Haiku 5.5.
| Workload | Cost | Tokens |
|---|---|---|
| Short chat turn | $0.00035 | 1K in / 500 out |
| Summarize a long document | $0.011 | 100K in / 2K out |
| Agentic coding session | $0.100 | 500K in / 100K out |
| 1M input + 1M output tokens | $0.600 | 1M in / 1M out |
Effective Cost
In AI IQ's scoring, Haiku 5.5's list price is adjusted by a token-usage multiplier of 0.477, giving an effective cost of $0.2865 per 1M input + output tokens. That places it #17 of 148 models on the effective-cost ranking (lower is cheaper). Some models spend far more tokens than others on the same task, so effective cost compares what a unit of work really costs. Method details are on the methodology page; see all models on the cost charts and the falling cost of intelligence over time.