Claude Haiku 4.5
- Input
- $1.00
- Output
- $5.00
- Cache read
- $0.100
- Model ID
- claude-haiku-4.5
Anthropic API Pricing Calculator
Estimate Anthropic API pricing and Claude API pricing for Claude model token costs using input tokens, output tokens, prompt caching, cached prompts, and monthly Claude API cost estimation assumptions.
Calculate monthly input and output token costs using current Anthropic model prices.
- MODELS COVERED
- 14
- DATA SOURCE
- Official Anthropic
- PRICING UNIT
- Per 1M tokens
- LAST VERIFIED
- Oct 9, 2026
API cost estimate
Estimate Claude API Cost
Choose a current Anthropic model from the same pricing rows, enter input tokens, output tokens, and request volume, then review the estimated monthly API cost.
Claude cost planning
Estimate Claude model token costs before long-context usage scales.
Start with the selected Claude model, input tokens, output tokens, prompt caching assumptions, average conversation length, and monthly request volume. Then compare the estimate against budget limits and other provider routes.
Anthropic API pricing
Current Claude API prices at a glance.
These Claude price cards use the same source-linked Anthropic rows as the full table below. Prices are shown in USD per 1M tokens, and Not listed values are not treated as zero.
Claude Sonnet 5
- Input
- $2.00
- Output
- $10.00
- Cache read
- $0.200
- Model ID
- claude-sonnet-5
Claude Opus 5
- Input
- $5.00
- Output
- $25.00
- Cache read
- $0.500
- Model ID
- claude-opus-5
Claude API monthly estimate
Example monthly Claude API costs.
Each example uses 1,000,000 input tokens + 250,000 output tokens and the same token-cost helper as the pricing dataset. Amounts are estimates only; actual Anthropic invoices depend on caching, retries, routing, account terms, and current official billing rules.
Claude Haiku 4.5
- Input tokens
- 1,000,000
- Output tokens
- 250,000
- Estimated monthly cost
- $2.25
- Model row
- claude-haiku-4.5
Claude Sonnet 5
- Input tokens
- 1,000,000
- Output tokens
- 250,000
- Estimated monthly cost
- $4.50
- Model row
- claude-sonnet-5
Claude Opus 5
- Input tokens
- 1,000,000
- Output tokens
- 250,000
- Estimated monthly cost
- $11.25
- Model row
- claude-opus-5
Claude prompt caching
How Anthropic prompt caching affects API cost.
Claude prompt caching separates normal input tokens, cache-write behavior, cache-read tokens, and output tokens. Repeated large system prompts, policy text, documents, or shared context may benefit when they qualify for Anthropic prompt caching and achieve useful cache hit rates.
Standard input
Input pricing applies to new prompt content sent to Claude, including messages, instructions, retrieved context, and long documents that are not billed as cache reads.
Repeated prompt context
- Claude Haiku 4.5
- $0.100
- Claude Sonnet 5
- $0.200
- Claude Opus 5
- $0.500
Cache write
Anthropic may bill cache writes separately under prompt caching rules, but the current provider rows do not expose a separate numeric cache-write field. Verify official Anthropic pricing before modeling cache-write charges or long-context caching savings.
Dedicated model pages
Pricing by model
Open a model page for its verified source, workload examples, price history, lifecycle details, and nearby alternatives.
Anthropic API pricing table
Anthropic API pricing table
Compare Claude input token, output token, and cached prompt pricing from the existing source-linked dataset. Not listed values are never treated as zero.
| Model | Input | Output | Cached Input | Batch Input | Batch Output | Unit | Status |
|---|---|---|---|---|---|---|---|
Claude Fable 5Model ID: claude-fable-5Provider: AnthropicDisplayed pricing: StandardEvidence | Input · USD per 1M input tokens$10.00 | Output · USD per 1M output tokens$50.00 | Cached input · USD per 1M cached input tokens$1.00 | Batch input · USD per 1M input tokens$5.00 | Batch output · USD per 1M output tokens$25.00 | USD per 1M tokens | activeStandard |
Claude Fable 5.1Model ID: claude-fable-5-1Provider: AnthropicDisplayed pricing: StandardEvidence Verified: Not availableChecked: Not availableOfficial sources (4): Official source 1Official source 2Official source 3Official source 4 | Input · USD per 1M input tokens$10.00 | Output · USD per 1M output tokens$50.00 | Cached input · USD per 1M cached input tokens$0.250 | Batch input · USD per 1M input tokens$5.00 | Batch output · USD per 1M output tokens$25.00 | USD per 1M tokens | latestStandard |
Claude Haiku 4.5Model ID: claude-haiku-4.5Provider: AnthropicDisplayed pricing: StandardEvidence | Input · USD per 1M input tokens$1.00 | Output · USD per 1M output tokens$5.00 | Cached input · USD per 1M cached input tokens$0.100 | Batch input · USD per 1M input tokens$0.500 | Batch output · USD per 1M output tokens$2.50 | USD per 1M tokens | latestStandard |
Claude Haiku 5.5Model ID: claude-haiku-5-5Provider: AnthropicDisplayed pricing: Standard · Short context · ≤100k prompt tokens · whole requestEvidence Verified: Checked: Official sources (4): Official source 1Official source 2Official source 3Official source 4 | Input · USD per 1M input tokens$0.100 | Output · USD per 1M output tokens$0.500 | Cached input · USD per 1M cached input tokens$0.010 | Batch input · USD per 1M input tokens$0.050 | Batch output · USD per 1M output tokens$0.250 | USD per 1M tokens | latestactive |
Claude Mythos 5Model ID: claude-mythos-5Provider: AnthropicDisplayed pricing: StandardEvidence | Input · USD per 1M input tokens$10.00 | Output · USD per 1M output tokens$50.00 | Cached input · USD per 1M cached input tokens$1.00 | Batch input · USD per 1M input tokensNot listed | Batch output · USD per 1M output tokensNot listed | USD per 1M tokens | limitedLimited availability |
Claude Mythos 5.1Model ID: claude-mythos-5-1Provider: AnthropicDisplayed pricing: StandardEvidence Verified: Not availableChecked: Not availableOfficial sources (3): Official source 1Official source 2Official source 3 | Input · USD per 1M input tokens$10.00 | Output · USD per 1M output tokens$50.00 | Cached input · USD per 1M cached input tokens$0.250 | Batch input · USD per 1M input tokens$5.00 | Batch output · USD per 1M output tokens$25.00 | USD per 1M tokens | limitedLimited availability |
Claude Opus 4.1Model ID: claude-opus-4.1Provider: AnthropicEvidence | Input · USD per 1M input tokens$15.00 | Output · USD per 1M output tokens$75.00 | Cached input · USD per 1M cached input tokens$1.50 | Batch input · USD per 1M input tokensNot listed | Batch output · USD per 1M output tokensNot listed | USD per 1M tokens | retiredRetired |
Claude Opus 4.8Model ID: claude-opus-4.8Provider: AnthropicDisplayed pricing: StandardEvidence | Input · USD per 1M input tokens$5.00 | Output · USD per 1M output tokens$25.00 | Cached input · USD per 1M cached input tokens$0.500 | Batch input · USD per 1M input tokens$2.50 | Batch output · USD per 1M output tokens$12.50 | USD per 1M tokens | latestStandard |
Claude Opus 5Model ID: claude-opus-5Provider: AnthropicDisplayed pricing: StandardEvidence | Input · USD per 1M input tokens$5.00 | Output · USD per 1M output tokens$25.00 | Cached input · USD per 1M cached input tokens$0.500 | Batch input · USD per 1M input tokens$2.50 | Batch output · USD per 1M output tokens$12.50 | USD per 1M tokens | latestStandard |
Claude Opus 5.5Model ID: claude-opus-5-5Provider: AnthropicDisplayed pricing: StandardEvidence | Input · USD per 1M input tokens$4.00 | Output · USD per 1M output tokens$20.00 | Cached input · USD per 1M cached input tokens$0.200 | Batch input · USD per 1M input tokensNot listed | Batch output · USD per 1M output tokensNot listed | USD per 1M tokens | latestactive |
Claude Sonnet 4Model ID: claude-sonnet-4Provider: AnthropicEvidence | Input · USD per 1M input tokens$3.00 | Output · USD per 1M output tokens$15.00 | Cached input · USD per 1M cached input tokens$0.300 | Batch input · USD per 1M input tokensNot listed | Batch output · USD per 1M output tokensNot listed | USD per 1M tokens | retiredRetired except partner clouds |
Claude Sonnet 4.6Model ID: claude-sonnet-4.6Provider: AnthropicDisplayed pricing: StandardEvidence | Input · USD per 1M input tokens$3.00 | Output · USD per 1M output tokens$15.00 | Cached input · USD per 1M cached input tokens$0.300 | Batch input · USD per 1M input tokens$1.50 | Batch output · USD per 1M output tokens$7.50 | USD per 1M tokens | latestStandard |
Claude Sonnet 5Model ID: claude-sonnet-5Provider: AnthropicDisplayed pricing: StandardEvidence Verified: Not availableChecked: Not availableOfficial sources (2): Official source 1Official source 2 | Input · USD per 1M input tokens$2.00 | Output · USD per 1M output tokens$10.00 | Cached input · USD per 1M cached input tokens$0.200 | Batch input · USD per 1M input tokens$1.00 | Batch output · USD per 1M output tokens$5.00 | USD per 1M tokens | latestStandard |
Claude Sonnet 5.5Model ID: claude-sonnet-5-5Provider: AnthropicDisplayed pricing: StandardEvidence | Input · USD per 1M input tokens$2.00 | Output · USD per 1M output tokens$10.00 | Cached input · USD per 1M cached input tokens$0.100 | Batch input · USD per 1M input tokens$1.00 | Batch output · USD per 1M output tokens$5.00 | USD per 1M tokens | latestactive |
Use these Anthropic pricing rows as planning inputs: calculate your monthly API cost with realistic token counts, or compare this provider with other AI models before routing production traffic.
For deeper pricing context, read GPT-6 and Claude Opus 5.5 API costs and how Claude Code subscription and API billing differ and Claude API starter credits and the cheapest legitimate path.
Calculate whether Anthropic prompt caching saves money for your workload using verified mode and lifetime data.
Canonical pricing tiers
Claude Haiku 5.5 short- and long-context pricing.
These standard API prices come from the canonical tier records attached to this model. Values are shown in USD per 1M tokens and are not inferred from the base row.
| Context tier | Total prompt token rule | Regular input | Cached input | Output | Unit |
|---|---|---|---|---|---|
| Short contextshort | ≤100k prompt tokens | $0.100 | $0.010 | $0.500 | USD / 1M tokens |
| Long contextlong | >100k prompt tokens | $0.500 | $0.050 | $2.50 | USD / 1M tokens |
Total prompt tokens
Total prompt tokens are regular input tokens plus cached input tokens. Both count toward the tier threshold.
Whole-request pricing
When a tier's threshold condition matches, that tier prices the entire standard request. This is a tier switch, not progressive pricing across token bands.
Price-record evidence
Claude Haiku 5.5
Model ID: claude-haiku-5-5
Provider: Anthropic
Evidence
Verified: Checked: Official sources (4): Official source 1Official source 2Official source 3Official source 4
Claude API cost calculator
Turn Claude token assumptions into a monthly cost estimate.
Use these planning modules with the calculator instead of treating the pricing table as a final bill. Anthropic API pricing depends on request shape, Claude model choice, prompt caching behavior, average conversation length, and production traffic.
Anthropic API pricing table
Review Claude model rows with input, output, cached prompt, unit, status, and official source links. The table below stays connected to the existing project pricing data instead of using hand-written Claude prices.
Claude API cost calculator
Move realistic input tokens, output tokens, requests per month, average conversation length, selected Claude model, and cached prompt assumptions into the calculator to estimate monthly Claude API costs.
Open the Claude API cost calculatorClaude model cost comparison
Compare Claude models by token price, context needs, output length, prompt caching fit, latency expectations, and quality requirements before routing production traffic.
Compare Claude model costsClaude token price examples
Common Claude API workloads to model.
These examples keep Claude token prices as variables. Add your own requests per month, input tokens, output tokens, selected Claude model, cached prompt usage, and average conversation length in the calculator.
Customer support chatbot using Claude
A Claude support chatbot estimate depends on requests per month, customer message size, retrieved help-center context, output tokens per answer, selected Claude model, prompt caching usage, and average conversation length.
Long document summarizer
A document summarizer is often input-heavy because long PDFs, contracts, research notes, or transcripts become Claude input tokens. Model document count, source length, summary length, selected Claude model, cached prompt usage, and follow-up turns.
SaaS AI assistant or coding assistant
A SaaS or coding assistant may include user messages, code snippets, product context, tool outputs, and multi-turn replies. Estimate requests per month, input tokens, output tokens, selected Claude model, prompt caching, and average conversation length.
Cheapest Claude API models
Find lower-cost Claude routes without losing task quality.
The cheapest Claude model is workload-specific. Compare total workflow cost, long-context size, answer quality, output length, prompt caching hit rate, retries, and fallback behavior before changing model routes.
Claude Haiku 5.5
- Input
- $0.100
- Output
- $0.500
- Cached prompt
- $0.010
Claude Haiku 4.5
- Input
- $1.00
- Output
- $5.00
- Cached prompt
- $0.100
Claude Sonnet 5
- Input
- $2.00
- Output
- $10.00
- Cached prompt
- $0.200
Cheapest Claude model cards are derived from the same pricing rows shown on this page and are informational only. Verify current Anthropic API pricing, prompt caching rules, billing dashboards, and invoices before final purchasing decisions.
How Anthropic API token pricing works
Separate prompt cost, generated text, and cached prompt context.
Input tokens
Input tokens include user messages, system instructions, conversation history, retrieved documents, tool context, and long prompt content sent to Claude.
Output tokens
Output tokens are generated by Claude. Long summaries, detailed support replies, code explanations, and multi-turn assistant responses can make output a major part of Claude API costs.
Prompt caching
Prompt caching can help when stable instructions, documents, or shared app context repeat across requests. Keep cacheable prompt sections consistent and verify current Anthropic caching rules.
Official sources
Source URLs used for this provider.
Cost examples
- 1M input tokens
- $10.00
- Model
- Claude Fable 5
- 1M output tokens
- $50.00
- Model
- Claude Fable 5
- 10M input + 2M output
- $200.00
- Model
- Claude Fable 5
Calculate Claude API costs with your own usage assumptions.
Use the calculator for monthly Claude app usage, long-context examples, prompt caching scenarios, and model-routing comparisons.
Pricing data is informational and may lag provider changes. Always verify final billing decisions against provider pricing pages, contracts, dashboards, and invoices.
Anthropic API pricing FAQ
How much does the Anthropic API cost?
Anthropic API cost depends on the selected Claude model, input tokens, output tokens, prompt caching behavior, request volume, average conversation length, retries, and monthly usage. Use the pricing table and calculator together to estimate your own workload.
How much does the Claude API cost?
Claude API cost is usage-based and varies by Claude model and token mix. A short support answer, a long document summary, and a coding assistant session can have very different costs because each sends different input tokens and generates different output tokens.
Is the Anthropic API free?
The Anthropic API is generally a paid API service. Any free credits, trials, grants, or account-specific terms should be checked in Anthropic's official console, pricing pages, and billing dashboard before you plan production usage.
How do input and output tokens affect Claude API pricing?
Input tokens are the prompt, user message, system instructions, chat history, retrieved documents, and tool context sent to Claude. Output tokens are generated by Claude. Since the two sides can be priced differently, long prompts and long answers should be estimated separately.
What is prompt caching in Claude API pricing?
Prompt caching can reduce the cost of repeated prompt content when stable context qualifies under Anthropic's caching rules. Product instructions, policy text, long documents, and common system prompts may benefit when they are reused consistently, but eligibility and final billing should be verified with official Anthropic documentation and invoices.
How can I estimate monthly Claude API costs?
Estimate one representative Claude request first: input tokens, output tokens, selected Claude model, average conversation length, and cached prompt usage. Then multiply by requests per month and add realistic retry, fallback, and high-usage assumptions.
Which Claude API model is cheapest?
The cheapest Claude API model depends on the current pricing table, task quality, output length, cache hit rate, context size, and retry rate. Smaller or faster Claude models can be lower cost for routine work, while larger models may be worth it for complex reasoning or long-context tasks.
Is Claude cheaper than OpenAI for long prompts?
Claude can be more or less expensive than OpenAI depending on model choice, long-context size, output length, prompt caching, and quality needs. For long prompts, compare total workflow cost rather than only input token price.