Claude API pricing

Oxyy is an independent API gateway, not affiliated with Anthropic.

Paid Claude text models run from $0.040 to $4.00 per 1M input tokens and $0.200 to $20.00 per 1M output tokens. Claude Opus 5 costs $2.00 input / $10.00 output per 1M tokens, pay-as-you-go (standard price $5.00 / $25.00). The lowest-priced paid Claude text model is Claude Haiku 5.5 at $0.040 input / $0.200 output per 1M tokens.

Claude API pricing per 1M tokens

Prices in USD per 1M tokens. Where two prices are shown, the first is the standard price and the second is what you pay on Oxyy, pay as you go. Cached input is the Oxyy price.
ModelContextInputOutputCached inputSavingReleased
Claude Opus 5claude-opus-51MStandard price $5.00, on Oxyy $2.00Standard price $25.00, on Oxyy $10.00$0.20060%—
Claude Sonnet 5claude-sonnet-51MStandard price $2.00, on Oxyy $0.800Standard price $10.00, on Oxyy $4.00$0.08060%—
Claude Haiku 4.5claude-haiku-4-5-20251001200KStandard price $1.00, on Oxyy $0.400Standard price $5.00, on Oxyy $2.00$0.04060%Oct 1, 2025
Claude Opus 5.5claude-opus-5-51MStandard price $4.00, on Oxyy $1.60Standard price $20.00, on Oxyy $8.00$0.08060%Sep 22, 2026
Claude Haiku 5.5 (long prompts cost more, see note)claude-haiku-5-51MStandard price $0.100, on Oxyy $0.040Standard price $0.500, on Oxyy $0.200$0.00460%Oct 7, 2026
Claude Sonnet 5.5claude-sonnet-5-51MStandard price $2.00, on Oxyy $0.800Standard price $10.00, on Oxyy $4.00$0.08060%Sep 28, 2026
Claude Sonnet 4.6claude-sonnet-4-61MStandard price $3.00, on Oxyy $1.20Standard price $15.00, on Oxyy $6.00$0.12060%Feb 17, 2026
Claude Fable 5claude-fable-51MStandard price $10.00, on Oxyy $4.00Standard price $50.00, on Oxyy $20.00$0.40060%Jun 9, 2026
Claude Opus 4.8claude-opus-4-81MStandard price $5.00, on Oxyy $2.00Standard price $25.00, on Oxyy $10.00$0.20060%—
Claude Sonnet 4.5claude-sonnet-4-5-20250929200KStandard price $3.00, on Oxyy $1.20Standard price $15.00, on Oxyy $6.00$0.12060%Sep 29, 2025
Claude Opus 4.6claude-opus-4-61MStandard price $5.00, on Oxyy $2.00Standard price $25.00, on Oxyy $10.00$0.20060%Feb 5, 2026
Claude Fable 5.1claude-fable-5-11MStandard price $10.00, on Oxyy $4.00Standard price $50.00, on Oxyy $20.00$0.10060%—
Claude Opus 4.7claude-opus-4-71MStandard price $5.00, on Oxyy $2.00Standard price $25.00, on Oxyy $10.00$0.20060%—
Claude Opus 4.5claude-opus-4-5-20251101200KStandard price $5.00, on Oxyy $2.00Standard price $25.00, on Oxyy $10.00$0.20060%Nov 24, 2025
  • † Long prompts cost more. The whole request is billed at the higher rate:
    • Claude Haiku 5.5: $0.200 input / $1.00 output per 1M tokens once a prompt goes over 100K tokens.
  • The standard price is Oxyy's list price for the model before any discount. Anthropic publishes its own Claude prices on its pricing page.

Claude API cost calculator

Estimate a month of Claude API usage: set your input and output volume and how much of your input is a repeated prompt prefix served from cache, or start from a preset.

Estimated monthly cost for 10M input and 2M output tokens, with 0% of input served from cache.

Estimates in USD per month. Standard price: Oxyy's list price before any discount. On Oxyy: the pay-as-you-go price you are billed. Taxes are not included.
Claude modelAt standard priceOn OxyyYou save
Claude Opus 5$100.00$40.00$60.00 (60%)
Claude Sonnet 5$40.00$16.00$24.00 (60%)
Claude Haiku 4.5$20.00$8.00$12.00 (60%)
Claude Opus 5.5$80.00$32.00$48.00 (60%)
Claude Haiku 5.5 (estimate is a lower bound, see note)$2.00$0.80$1.20 (60%)
Claude Sonnet 5.5$40.00$16.00$24.00 (60%)
Claude Sonnet 4.6$60.00$24.00$36.00 (60%)
Claude Fable 5$200.00$80.00$120.00 (60%)
  • † A lower bound: prompts past the model's long-context threshold are billed at a higher rate (see the table above).

Which Claude model should you use?

For each kind of work, the lowest-priced paid Claude model tagged for it and the one developers use most on Oxyy.

Lowest price: the paid model tagged for the need with the lowest Oxyy price (3 input tokens to 1 output token). Most used: the busiest of those models on Oxyy over the last 7 days.
NeedLowest priceMost used on Oxyy
CodingClaude Sonnet 5.5$0.800 / $4.00 per 1M tokensClaude Opus 5.5
ReasoningClaude Sonnet 5.5$0.800 / $4.00 per 1M tokensClaude Opus 5.5
Agents and tool useClaude Sonnet 5.5$0.800 / $4.00 per 1M tokensClaude Opus 5.5
Long documentsClaude Haiku 5.5$0.040 / $0.200 per 1M tokensClaude Opus 5.5
Lowest costClaude Haiku 5.5$0.040 / $0.200 per 1M tokensClaude Opus 5.5

Claude API pricing FAQ

How much does the Claude API cost?

On Oxyy, Claude Opus 5 costs $2.00 input / $10.00 output per 1M tokens at the pay-as-you-go price (standard price $5.00 / $25.00). Across paid Claude text models, input runs from $0.040 to $4.00 per 1M tokens and output from $0.200 to $20.00. There is no subscription: you pay only for what you use.

What is the cheapest Claude model?

Claude Haiku 5.5 is the lowest-priced paid Claude text model on Oxyy, at $0.040 input / $0.200 output per 1M tokens.

How much does cached input cost on Claude models?

Cached input on Claude Opus 5 costs $0.200 per 1M tokens on Oxyy, against $2.00 for regular input. Writing to the cache costs $2.50 per 1M tokens for a 5-minute cache and $4.00 for a 1-hour cache. The price table on this page lists the cache rate of every model that has one.

Can I call Claude models with the OpenAI SDK?

Yes. Oxyy's API is OpenAI-compatible: set the base URL to https://api.oxyy.ai/v1, use your Oxyy API key and pass the model id, for example "claude-opus-5", to /v1/chat/completions.

Do long prompts cost more on Claude models?

For some models, yes. Claude Haiku 5.5 costs $0.200 input / $1.00 output per 1M tokens once a prompt goes over 100K tokens, instead of $0.040 / $0.200; the whole request is priced at the higher rate.

Is Oxyy affiliated with Anthropic?

No. Oxyy is an independent API gateway, not affiliated with or endorsed by Anthropic. It gives you one API key and one bill for models from many AI labs, Anthropic among them.

All Anthropic models 14

Anthropic: Claude Opus 5.5TextImageText-60%697.9M tokens
reasoningcodingagenticlong-contextflagship

Built for long-running agentic coding and knowledge work. Anthropic's current default flagship.

by AnthropicSep 22, 20261M context$1.60/M input standard price $4.00$8.00/M output standard price $20.0012.8s latency183 t/s
Anthropic: Claude Haiku 5.5TextImageText-60%Haiku492.1M tokens
fastcheapclassificationroutingextraction+2 more

Anthropic's fastest model, for high-volume latency-sensitive work: classification, routing, extraction and subagent tasks. 1M context with adaptive thinking.

by AnthropicOct 7, 20261M context$0.040/M input standard price $0.100$0.200/M output standard price $0.5003954ms latency310 t/s
Anthropic: Claude Sonnet 5.5TextImageText-60%176.4M tokens
reasoningcodingagenticlong-contextfast

The best combination of speed and intelligence in the Claude lineup. Second model in the Claude 5.5 family.

by AnthropicSep 28, 20261M context$0.800/M input standard price $2.00$4.00/M output standard price $10.009370ms latency248 t/s
Anthropic: Claude Sonnet 5TextImageText-60%76.7M tokens
reasoningcodingagenticlong-context

The best combination of speed and intelligence in the Claude lineup.

by Anthropic1M context$0.800/M input standard price $2.00$4.00/M output standard price $10.003653ms latency148 t/s
Anthropic: Claude Opus 5TextImageText-60%10.7M tokens
reasoningcodingagenticlong-context

For complex agentic coding and enterprise work. A step-change over Claude Opus 4.8 in deep reasoning, agentic and long-horizon tasks, and test-time compute scaling.

by Anthropic1M context$2.00/M input standard price $5.00$10.00/M output standard price $25.007462ms latency138 t/s
Anthropic: Claude Haiku 4.5TextImageText-60%10.1M tokens
fastlow-costclassificationhigh-volume

The fastest Claude model with near-frontier intelligence.

by AnthropicOct 1, 2025200K context$0.400/M input standard price $1.00$2.00/M output standard price $5.001630ms latency108 t/s
Anthropic: Claude Sonnet 4.6TextImageText-60%6.9M tokens
reasoningcodingagenticlong-context

Legacy Sonnet model. The first Sonnet with the full 1M token context window at standard pricing.

by AnthropicFeb 17, 20261M context$1.20/M input standard price $3.00$6.00/M output standard price $15.004019ms latency63 t/s
Anthropic: Claude Fable 5TextImageText-60%2.6M tokens
reasoningcodingagenticlong-context

Previous-generation Fable-tier model for demanding reasoning and long-horizon agentic work. Superseded by Claude Fable 5.1.

by AnthropicJun 9, 20261M context$4.00/M input standard price $10.00$20.00/M output standard price $50.0022.8s latency273 t/s
Anthropic: Claude Opus 4.8TextImageText-60%2.4M tokens
reasoningcodingagenticlong-context

Previous-generation Opus model with adaptive thinking. Anthropic recommends migrating to Claude Opus 5.

by Anthropic1M context$2.00/M input standard price $5.00$10.00/M output standard price $25.003298ms latency90 t/s
Anthropic: Claude Sonnet 4.5TextImageFileText-60%1.8M tokens
reasoningcodingagentic

Legacy Sonnet model with a 200K token context window and extended thinking.

by AnthropicSep 29, 2025200K context$1.20/M input standard price $3.00$6.00/M output standard price $15.002339ms latency182 t/s
Anthropic: Claude Opus 4.6TextImageText-60%1.2M tokens
reasoningcodingagenticlong-context

Legacy Opus model. The first Opus with the full 1M token context window at standard pricing.

by AnthropicFeb 5, 20261M context$2.00/M input standard price $5.00$10.00/M output standard price $25.003217ms latency67 t/s
Anthropic: Claude Fable 5.1TextImageText-60%395.9K tokens
reasoningcodingagenticlong-context

Anthropic's most capable model, for demanding reasoning and long-horizon agentic work.

by Anthropic1M context$4.00/M input standard price $10.00$20.00/M output standard price $50.0010.3s latency87 t/s
Anthropic: Claude Opus 4.7TextImageText-60%217.9K tokens
reasoningcodingagenticlong-context

Legacy Opus model with adaptive thinking and the xhigh effort level for long-running agentic and coding tasks.

by Anthropic1M context$2.00/M input standard price $5.00$10.00/M output standard price $25.0010.6s latency89 t/s
Anthropic: Claude Opus 4.5TextImageText-60%5.7K tokens
reasoningcodingagentic

Legacy Opus model with extended thinking and a 200K context window.

by AnthropicNov 24, 2025200K context$2.00/M input standard price $5.00$10.00/M output standard price $25.003367ms latency154 t/s