Model Selection
Usage Parameters
Percentage of input tokens that are cached (significantly cheaper).
Estimated Monthly Cost
*Estimates based on current Claude API pricing. Actual costs may vary.
Complete Guide to Claude API Costs
Anthropic's Claude API pricing is based on model intelligence and speed. The family includes Haiku (fast & cheap), Sonnet (balanced), and Opus (most powerful). Choosing the right model is a critical part of any automation strategy.
Key Takeaways
- ●Claude 3.5 Sonnet is the best all-rounder for coding and reasoning.
- ●Claude 3.5 Haiku is faster and cheaper, ideal for simple tasks.
- ●Prompt Caching offers a 90% discount on cached input tokens.
Claude API Pricing Overview
| Model | Input Price ($/1M) | Output Price ($/1M) | Cached Input ($/1M) |
|---|---|---|---|
| Claude 4.5 Series (Latest) | |||
| Claude 4.5 Opus | $5.00 | $25.00 | $0.50 |
| Claude 4.5 Sonnet | $3.00 | $15.00 | $0.30 |
| Claude 4.5 Haiku | $1.00 | $5.00 | $0.10 |
| Claude 3.5 Series | |||
| Claude 3.5 Sonnet | $3.00 | $15.00 | $0.30 |
| Claude 3.5 Haiku | $1.00 | $5.00 | $0.10 |
| Claude 3 Series (Legacy) | |||
| Claude 3 Opus | $15.00 | $75.00 | $1.50 |
| Claude 3 Haiku | $0.25 | $1.25 | $0.03 |
Claude 3.5 Sonnet: The Best All-Rounder
Claude 3.5 Sonnet is widely considered the best model for coding and reasoning tasks, offering performance exceeding GPT-4o at a lower price point. It's the engine behind many advanced custom workflows.
Prompt Caching: A Game Changer
Anthropic's Prompt Caching allows you to cache large contexts (like documentation or codebases) for up to 5 minutes (refreshable). Cached input tokens are 90% cheaper than standard input tokens, making it feasible to build context-heavy applications.
- Write (Cache Creation): Standard input price + 25% surcharge.
- Read (Cache Hit): 90% discount on input price.
Claude API Pricing FAQs
Got questions? We've got answers. Here are the most common questions we get from potential clients.