Haiku 4.5 API Pricing
Haiku 4.5 has tracked API rates of $1.00 per 1M input tokens and $5.00 per 1M output tokens. Cached input is billed at $0.1 per 1M tokens. Rates are in USD from the AI IQ dataset; see the Haiku 4.5 model profile for capability scores.
Price List
| Rate | Price (USD) | Notes |
|---|---|---|
| Input | $1.00 | per 1M tokens |
| Output | $5.00 | per 1M tokens |
| Cached input (read) | $0.1 | per 1M tokens |
| Cache write | $1.25 | per 1M tokens |
| Cache write (1-hour) | $2.00 | per 1M tokens |
| Blended input + output | $6.00 | 1M in + 1M out at list price |
| Context window | 200K 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.
No model-specific official reference is recorded for this entry. Confirm the current rates directly with your inference provider.
What a Workload Costs
Computed straight from the list prices above — token counts are illustrative workload sizes, not measurements of Haiku 4.5.
| Workload | Cost | Tokens |
|---|---|---|
| Short chat turn | $0.00350 | 1K in / 500 out |
| Summarize a long document | $0.110 | 100K in / 2K out |
| Agentic coding session | $1.00 | 500K in / 100K out |
| 1M input + 1M output tokens | $6.00 | 1M in / 1M out |
Effective Cost
In AI IQ's scoring, Haiku 4.5's list price is adjusted by a token-usage multiplier of 0.5 (measured from real benchmark runs), giving an effective cost of $3.00 per 1M input + output tokens. That places it #73 of 153 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.
Cheaper Alternatives at Similar Capability
Models with a lower effective cost that score within a few IQ points of Haiku 4.5 (IQ 100), or better.
| Model | Provider | IQ | Effective Cost / 1M |
|---|---|---|---|
| DeepSeek V4.1 Flash | DeepSeek | 126 | $1.87 |
| gpt-6-luna | OpenAI | 125 | $0.7693 |
| mimo-v2.6-pro | Xiaomi | 125 | $1.52 |
| gpt-5.6-luna | OpenAI | 124 | $1.63 |
| mimo-v2.6-flash | Xiaomi | 122 | $0.6897 |
| deepseek-v4-flash | DeepSeek | 121 | $1.69 |