Prices checked 2026-10-02
Claude API Pricing (October 2026)
Claude API pricing runs from $1 per 1M input tokens on Claude Haiku 4.5 to $50 per 1M output tokens on Claude Fable 5.1. Claude Opus 5.5 costs $4 input and $20 output per 1M tokens. Output tokens cost 5x more than input, and batch processing halves both.
Current Claude models
Prices per 1 million tokens. Cached input is the price of reading a cached prompt prefix.
| Model | Input / 1M | Cached input | Output / 1M | Batch in / out | Context |
|---|---|---|---|---|---|
| Claude Fable 5.1 | $10 | $0.25 | $50 | $5 / $25 | 1M |
| Claude Opus 5 | $5 | $0.50 | $25 | $2.50 / $12.50 | 1M |
| Claude Opus 5.5 | $4 | $0.20 | $20 | $2 / $10 | 1M |
| Claude Sonnet 5 | $2 | $0.20 | $10 | $1 / $5 | 1M |
| Claude Haiku 4.5 | $1 | $0.10 | $5 | $0.50 / $2.50 | 200K |
What things actually cost
Four common jobs, priced on every current Claude model.
| Job | Claude Haiku 4.5 | Claude Sonnet 5 | Claude Opus 5.5 | Claude Opus 5 | Claude Fable 5.1 |
|---|---|---|---|---|---|
| Summarize a 2,000-word article2,667 input tokens in, a 300-word summary (400 tokens) out; total for 1,000 articles | $4.67 | $9.33 | $18.67 | $23.34 | $46.67 |
| Support chatbot, 10,000 messages a dayeach reply reads ~1,000 tokens of conversation and writes ~200 tokens; total for 30 days | $600 | $1,200 | $2,400 | $3,000 | $6,000 |
| Review a 5,000-line codebase50,000 input tokens at ~10 tokens per line, a 1,500-word review out; total for one review | $0.06 | $0.12 | $0.24 | $0.30 | $0.60 |
| Read a 300-page book and answer questions200,100 input tokens at ~667 tokens per page, 10 answers of ~200 words; total for one book | $0.21 | $0.43 | $0.85 | $1.07 | $2.13 |
Standard (non-batch, uncached) prices. Token counts use 1 token ≈ 0.75 English words; your real counts depend on each model's tokenizer.
Calculate your own budget
Pick a model and a budget to see how many tokens, words and pages it buys.
500,000tokens
Discounts and pricing rules
- Cache hits cost 0.1x the base input price — 0.05x on Claude Opus 5.5 and 0.025x on Claude Fable 5.1.
- The Batch API takes 50% off both input and output for asynchronous jobs.
- Claude 4.6 and later models bill the full 1M-token context window at standard rates — a 900K-token prompt costs the same per token as a 9K one.
- US-only inference (inference_geo: "us") adds a 1.1x multiplier on Claude 4.6 and later models.
- Fast mode (research preview, Claude API only) costs $8 / $40 per 1M tokens on Opus 5.5 and $10 / $50 on Opus 5 and Opus 4.8.
- Web search is billed at $10 per 1,000 searches on top of tokens.
Source: Anthropic's official pricing page. Base prices are re-checked daily against live market data.
Previous-generation Claude models
Still available through the API. Each row names its current replacement.
| Model | Input / 1M | Cached input | Output / 1M | Batch in / out | Context | Replaced by |
|---|---|---|---|---|---|---|
| Claude Fable 5 | $10 | $1 | $50 | $5 / $25 | 1M | Claude Fable 5.1 |
| Claude Opus 4.8 | $5 | $0.50 | $25 | $2.50 / $12.50 | 1M | Claude Opus 5.5 |
Frequently asked questions
- How much does the Claude API cost?
- Claude API prices run from $1 per 1M input tokens to $50 per 1M output tokens, depending on the model. Current models (input / output per 1M tokens): Claude Fable 5.1 $10 / $50, Claude Opus 5 $5 / $25, Claude Opus 5.5 $4 / $20, Claude Sonnet 5 $2 / $10, Claude Haiku 4.5 $1 / $5.
- What is the cheapest Claude model?
- Claude Haiku 4.5 at $1 per 1M input tokens and $5 per 1M output tokens, or $0.50 / $2.50 through the Batch API. One million input tokens is about 750,000 words.
- How much does Claude Opus 5.5 cost?
- Claude Opus 5.5 costs $4 per 1M input tokens and $20 per 1M output tokens, with a 1M-token context window. The most expensive Claude model is Claude Fable 5.1 at $10 / $50.
- How can I lower my Claude API bill?
- Cache hits cost 0.1x the base input price — 0.05x on Claude Opus 5.5 and 0.025x on Claude Fable 5.1. The Batch API takes 50% off both input and output for asynchronous jobs.
- Is the Claude API the same as a subscription?
- Claude.ai subscriptions (Free, Pro, Max, Team, Enterprise) are billed separately from the API. The API is pay-per-token with no monthly fee; see claude.com/pricing for subscription plans.