Anthropic Logo

Anthropic API Pricing Calculator

Anthropic is a leading AI research company building advanced multimodal and reasoning-focused AI models, known for Claude, delivering high-performance, safe, and scalable AI systems for enterprise and developer applications worldwide.

Anthropic’s Claude models enable developers to build reliable AI applications for coding, analysis, and automation while optimizing API usage. On ModelCosts.com, users can compare Claude pricing tiers and get pricing change alerts to maintain predictable spend across different workloads and deployment scales.

Last updated Oct 27, 2026

Never miss an AI pricing or launch update

Get notified about new model launches, pricing changes, provider updates, deprecations, and cost-saving insights.

🔒 We respect your privacy. Unsubscribe anytime.

Provider Details

Uptime Status
  • Headquarter: San Francisco, California, USA
  • Notable Models: Claude Opus, Claude Sonnet, Claude Haiku

Model Release Timeline

Anthropic develops cutting-edge AI models under the Claude family, focused on safety, reasoning, and long-context understanding. Its models, including Claude 3 Opus, Sonnet, and Haiku, support multimodal inputs and are widely used for coding, document analysis, and enterprise automation. Available via API and cloud platforms like AWS Bedrock and Google Cloud, Anthropic enables scalable AI deployment for businesses and developers. Claude models are particularly strong in structured reasoning and large-context tasks. Through ModelCosts.com, users can monitor Anthropic API pricing, compare model efficiency, and optimize costs with real-time insights, ensuring better control over AI infrastructure spending and performance trade-offs.
Selling Points:
- Advanced reasoning capabilities with strong performance in coding, analysis, and long-context tasks
- Multimodal support across text and vision inputs for diverse enterprise use cases
- Tiered model lineup (Opus, Sonnet, Haiku) optimized for performance vs. cost trade-offs
- API access via Anthropic, AWS Bedrock, and Google Cloud for flexible deployment
- Predictable and structured pricing suited for scaling production workloads
- Strong focus on AI safety and alignment, appealing to enterprise adoption
- Ideal for high-context workflows like legal, research, and technical documentation
- ModelCosts.com enables pricing comparisons, cost forecasting, and price change alerts for Anthropic APIs
- Competitive alternative to OpenAI and Google in premium AI model segment

Quick Model Estimate

(USD 0.1000 per 1M tokens)
(USD 0.5000 per 1M tokens)

Your Cost Estimate

💰 Total Cost

—

for 1000 input + 1000 output tokens

📥 Input (1000 × $0.100000) —
📤 Output (1000 × $0.500000) —

Cost Breakdown

📥 Input 📤 Output

Prices are watched for changes and checked against the provider's own page.

Pricing

2 statuses
Status
All categories
Category

Showing 16 of 23 models

Model ↕
Status ↕
Modality ↕
Input Price
(per 1M tokens) ↕
Output Price
(per 1M tokens) ↕
Cached Input
(per 1M tokens) ↕
Context Window ↕
View
Claude Haiku 4.5ActiveText$1.0000$5.0000$0.1000200,000 tokens→
Claude Opus 4.5ActiveText$5.0000$25.0000$0.5000200,000 tokens→
Claude Opus 4.6ActiveText$5.0000$25.0000$0.5000200,000 tokens→
Claude Sonnet 4.5Retiring SoonText$3.0000$15.0000$0.3000200,000 tokens→
Claude Sonnet 4.6ActiveText$3.0000$15.0000$0.3000200,000 tokens→
Claude Opus 4.7ActiveText$5.0000$25.0000$0.5000200,000 tokens→
Claude Fable 5ActiveText$10.0000$50.0000$1.00001,000,000 tokens→
Claude Mythos 5ActiveText$10.0000$50.0000$1.00001,000,000 tokens→
Claude Opus 4.8ActiveText$5.0000$25.0000$0.50001,000,000 tokens→
Claude Sonnet 5ActiveText$2.0000$10.0000$0.20001,000,000 tokens→
Claude Opus 5ActiveText$5.0000$25.0000$0.50001,000,000 tokens→
Claude Fable 5.160-min cacheActiveText$10.0000$50.0000$0.25001,000,000 tokens→
Claude Mythos 5.1ActiveText$10.0000$50.0000$0.25001,000,000 tokens→
Claude Sonnet 5.5ActiveText$2.0000$10.0000$0.10001,000,000 tokens→
Claude Opus 5.5ActiveText$4.0000$20.0000$0.20001,000,000 tokens→
Claude Haiku 5.5ActiveText$0.1000$0.5000$0.01001,000,000 tokens→

Standard prices for the first context band. Batch, other service tiers and long-context prices are on each model's Anthropic page.

Latest Models by Anthropic

Claude Haiku 5.5 Active
Reasoning
Oct 2026 1,000,000 ctx
Claude Sonnet 5.5 Active
Multimodal
Sep 2026 1,000,000 ctx
Claude Opus 5.5 Active
Multimodal
Sep 2026 1,000,000 ctx
Claude Mythos 5.1 Active
Multimodal
Sep 2026 1,000,000 ctx
Claude Fable 5.1 Active
Multimodal
Sep 2026 1,000,000 ctx

Recently Deprecated Models

Claude Sonnet 4.5 Legacy
Multimodal
Deprecated Nov 2026
Claude Opus 4.1 Deprecated
Multimodal
Deprecated Aug 2026
Claude Sonnet 4 Deprecated
Multimodal
Deprecated Jun 2026
Claude Opus 4 Deprecated
Multimodal
Deprecated Jun 2026
Claude Haiku 3 Deprecated
Multimodal
Deprecated Apr 2026

Recent changes at Anthropic

The latest price changes, releases, new listings and retirements involving Anthropic.

  1. Price change

    Anthropic cut the price of Claude Sonnet 5.5: Cached input $0.20 → $0.10 per 1M tokens (−50%).

  2. Tier price

    Anthropic cut the Batch price of Claude Sonnet 5.5: Cached input $0.10 → $0.05 per 1M tokens (−50%).

  3. New model

    Anthropic released Claude Haiku 5.5.

  4. Legacy

    Anthropic moved Claude Sonnet 4.5 to legacy, retiring it on 30th Nov 2026, with Claude Sonnet 5.5 as the suggested replacement.

  5. New model

    Anthropic released Claude Sonnet 5.5.

See all changes, new releases and retirements →

Compare Anthropic models

Head-to-head pages for models Anthropic owns — against each other, and against other providers.

Claude Sonnet 5.5 vs Claude Opus 5.5 Two Anthropic models Full comparison → Claude Haiku 5.5 vs Claude Sonnet 5.5 Two Anthropic models Full comparison → Claude Haiku 5.5 vs Claude Opus 5.5 Two Anthropic models Full comparison → Claude Opus 5.5 vs Claude Opus 5 Two Anthropic models Full comparison → Gemini 4 Argon vs Claude Sonnet 5.5 Against Google Full comparison → Gemini 4 Argon vs Claude Opus 5.5 Against Google Full comparison →

Every comparison we publish →