AI pricing · November 2026
Claude API pricing
| Provider | Cached in | Context | ||||
|---|---|---|---|---|---|---|
| Claude Haiku 5.5 prompts up to 100K tokens; over 100K: $0.50 in, $2.50 out, $0.05 cached. Batch 50% off | Anthropic | $0.100 | $0.500 | $0.010 | — | $2.00 |
| Claude Haiku 4.5 batch 50% off; cached = cache hits (5m write 1.25x input) | Anthropic | $1.00 | $5.00 | $0.100 | — | $20.00 |
| Claude Sonnet 5.5 batch 50% off; cached = cache hits (5m write 1.25x input) | Anthropic | $2.00 | $10.00 | $0.100 | 1M | $40.00 |
| Claude Sonnet 5 batch 50% off; cached = cache hits (5m write 1.25x input) | Anthropic | $2.00 | $10.00 | $0.200 | 1M | $40.00 |
| Claude Sonnet 4.6 batch 50% off; cached = cache hits (5m write 1.25x input) | Anthropic | $3.00 | $15.00 | $0.300 | 1M | $60.00 |
| Claude Opus 5.5 batch 50% off; cached = cache hits (5m write 1.25x input) | Anthropic | $4.00 | $20.00 | $0.200 | 1M | $80.00 |
| Claude Opus 5 batch 50% off; cached = cache hits (5m write 1.25x input) | Anthropic | $5.00 | $25.00 | $0.500 | 1M | $100 |
| Claude Fable 5.1 batch 50% off; cached = cache hits (5m write 1.25x input) | Anthropic | $10.00 | $50.00 | $0.250 | 1M | $200 |
Cached input and batch discounts
| Model | Input $/1M | Cached input $/1M | Cache saving | Batch |
|---|---|---|---|---|
| Claude Fable 5.1 | $10 | $0.25 | 98% off | 50% off |
| Claude Opus 5.5 | $4 | $0.20 | 95% off | 50% off |
| Claude Sonnet 5.5 | $2 | $0.10 | 95% off | 50% off |
| Claude Haiku 5.5 | $0.10 | $0.01 | 90% off | 50% off |
| Claude Opus 5 | $5 | $0.50 | 90% off | 50% off |
| Claude Sonnet 5 | $2 | $0.20 | 90% off | 50% off |
| Claude Sonnet 4.6 | $3 | $0.30 | 90% off | 50% off |
| Claude Haiku 4.5 | $1 | $0.10 | 90% off | 50% off |
Writing a prompt to the 5-minute cache costs 1.25x the base input price (2x for the 1-hour cache); reading it back costs the cached price.
What a typical workload costs
An app that sends 1,000 requests a day, each with a 2,000-token prompt and a 500-token answer, uses 60M input and 15M output tokens in 30 days. If 1,500 of those prompt tokens are the same system prompt and context every time, prompt caching bills 45M of them at the cached rate (cache writes and storage not included). Batch is the same traffic sent as asynchronous batch jobs.
| Model | Per month | With caching | With batch |
|---|---|---|---|
| Claude Fable 5.1 | $1,350 | $911 | $675 |
| Claude Opus 5.5 | $540 | $369 | $270 |
| Claude Sonnet 5.5 | $270 | $185 | $135 |
| Claude Haiku 5.5 | $13.50 | $9.45 | $6.75 |
| Claude Opus 5 | $675 | $473 | $338 |
| Claude Sonnet 5 | $270 | $189 | $135 |
| Claude Sonnet 4.6 | $405 | $284 | $203 |
| Claude Haiku 4.5 | $135 | $94.50 | $67.50 |
Closest-priced models from other providers
Matched on a 3:1 mix of input to output tokens. Open-model hosts are left out.
| Anthropic model | In / out $/1M | Closest elsewhere |
|---|---|---|
| Claude Fable 5.1 | $10 / $50 | GPT-6 Astra (OpenAI) $10 / $50 · Gemini 3.1 Pro Preview (Google) $2 / $12 |
| Claude Opus 5.5 | $4 / $20 | GPT-5.3 Codex (OpenAI) $1.75 / $14 · Gemini 3.1 Pro Preview (Google) $2 / $12 |
| Claude Sonnet 5.5 | $2 / $10 | GPT-6.1 Sol (OpenAI) $2 / $10 · Gemini 4 Argon (Google) $2 / $10 |
| Claude Haiku 5.5 | $0.10 / $0.50 | GPT-6 Luna (OpenAI) $0.10 / $0.50 · Ministral 3 14B (Mistral) $0.20 / $0.20 |
| Claude Opus 5 | $5 / $25 | GPT-6 Astra (OpenAI) $10 / $50 · Gemini 3.1 Pro Preview (Google) $2 / $12 |
| Claude Sonnet 5 | $2 / $10 | GPT-6.1 Sol (OpenAI) $2 / $10 · Gemini 4 Argon (Google) $2 / $10 |
| Claude Sonnet 4.6 | $3 / $15 | GPT-5.3 Codex (OpenAI) $1.75 / $14 · Gemini 3.1 Pro Preview (Google) $2 / $12 |
| Claude Haiku 4.5 | $1 / $5 | GPT-6.1 Terra (OpenAI) $1 / $5 · DeepSeek V4 Pro (DeepSeek) $1.32 / $3.96 |
Source: Anthropic pricing, checked 2026-11-02. To price a real prompt, paste it into the token counter.
FAQ
How much does the Claude API cost?+
Anthropic charges per token. On 2026-11-02 its official pricing page listed, per 1M tokens: Claude Fable 5.1 $10 input / $50 output, Claude Opus 5.5 $4 input / $20 output, Claude Sonnet 5.5 $2 input / $10 output, Claude Haiku 5.5 $0.10 input / $0.50 output and Claude Opus 5 $5 input / $25 output, plus 3 more in the table above.
What is the cheapest Anthropic model?+
Claude Haiku 5.5: $0.10 per 1M input tokens and $0.50 per 1M output tokens. For the example workload on this page (30,000 requests a month) that is about $13.50 a month, against $1,350 on Claude Fable 5.1, the most expensive Anthropic model here.
Does Anthropic have prompt caching or cheaper cached input?+
Yes. Cached input tokens cost $0.25 on Claude Fable 5.1 (98% less than regular input), $0.20 on Claude Opus 5.5 (95% less than regular input), $0.10 on Claude Sonnet 5.5 (95% less than regular input) and $0.01 on Claude Haiku 5.5 (90% less than regular input). Writing a prompt to the 5-minute cache costs 1.25x the base input price (2x for the 1-hour cache); reading it back costs the cached price.
Does Anthropic offer a batch discount?+
Yes: 50% off for asynchronous batch requests on all 8 models here. Batch requests are processed in the background instead of answered right away, so they suit evals, backfills and bulk summaries rather than chat.
How does Claude Fable 5.1 compare on price with other AI APIs?+
At $10 input / $50 output per 1M tokens, the closest-priced models from other providers are GPT-6 Astra (OpenAI, $10 / $50) and Gemini 3.1 Pro Preview (Google, $2 / $12). Price per token isn't the whole bill: compare quality on your own prompts, and how many tokens each model uses for the same task.
How current are these prices?+
They were copied from Anthropic's official pricing page on 2026-11-02, and the month in this page's title is that check date. Prices change often, so confirm on the official page before you commit to a budget.
Other providers
OpenAI API pricing · Gemini API pricing · Grok API pricing · Mistral API pricing · DeepSeek API pricing · Groq API pricing · Together AI API pricing · Fireworks AI API pricing · All AI model prices compared