API pricing / November 2026
Claude and ChatGPT API pricing calculator
See what your monthly AI bill should be, and how much caching and batch processing can take off it.
Current API prices
Prices checked on 21 November 2026 against the official price lists: Claude pricing · OpenAI pricing
Anthropic
| Model | Input | Cached input | Output |
|---|---|---|---|
| Claude Fable 5.2 | $8.00 | $0.20 | $40.00 |
| Claude Opus 5.5 | $4.00 | $0.20 | $20.00 |
| Claude Sonnet 5.5 | $2.00 | $0.10 | $10.00 |
| Claude Haiku 5.5Prompts up to 100,000 tokens | $0.10 | $0.01 | $0.50 |
OpenAI
| Model | Input | Cached input | Output |
|---|---|---|---|
| gpt-6.2-astra | $8.00 | $0.40 | $40.00 |
| gpt-6.2-sol | $2.00 | $0.10 | $10.00 |
| gpt-6-astra | $10.00 | $1.00 | $50.00 |
| gpt-6.1-sol | $2.00 | $0.10 | $10.00 |
USD per 1M tokens. Batch processing: 50% off input and output at both providers.
Estimate your monthly cost
Enter input and output tokens to see the estimate.
Cache write surcharges are ignored in this estimate.
Do not move to a smaller model on price alone. Compare results on a sample of your own requests first.
API pricing questions
- How does API pricing work?
- Providers charge for the tokens your application sends and receives. Input and output have separate prices, usually stated in US dollars per million tokens. Cached input has its own lower price, and eligible batch work receives a discount.
- What is a token?
- A token is a unit of text used by a language model. It can be a whole word, part of a word or punctuation. Your input tokens include the instructions and context sent to the model; output tokens are the response it generates.
- What is cached input?
- Cached input is repeated context that the provider can reuse instead of processing as new input every time. It typically contains stable instructions, reference material or other context shared across requests. The provider's usage data shows how much input was read from cache.
- What is batch processing?
- Batch processing lets you submit work that does not need an immediate response for later processing. It suits asynchronous jobs rather than live interactions. In return, both providers listed here offer 50% off input and output tokens.