Build with Claude

Experiment, implement, and scale with the Claude API. Choose from the following account types.

  • Build

    Create a proof-of-concept and launch your own generative AI solution.

    On the Build plan, you get:

    Get API Access
  • Scale

    Scale your generative AI solution with custom rate limits and hands-on support from the Anthropic team.

    On the Scale plan, you get:

    • All the benefits of the Build plan
    • Anthropic-supported onboarding
    • Custom rate limits
    • Billing via monthly invoices
    • Access to prompting and deployment support
    Contact Sales

Claude 3 Pricing

MTok = million tokens. All Claude 3 models support vision and 200,000 token context windows.
Light & fast

Haiku

  • Input: $0.25 / MTok
  • Output: $1.25 / MTok
Hard-working

Sonnet

  • Input: $3 / MTok
  • Output: $15 / MTok
Powerful

Opus

  • Input: $15 / MTok
  • Output: $75 / MTok
Cost
Intelligence

Legacy Model Pricing

  • Claude 2.1

    200,000 token context window

    • Input: $8 / MTok
    • Output: $24 / MTok
  • Claude 2.0

    100,000 token context window

    • Input: $8 / MTok
    • Output: $24 / MTok
  • Claude Instant

    100,000 token context window

    • Input: $0.80 / MTok
    • Output: $2.40 / MTok

Partners building with Claude

Asana Logo
Assembly AI Logo
BCG Logo
Brave Logo
Bridgewater
DuckDuckGo Logo
GitLab Logo
Jane Street Logo
Jasper Logo
LexisNexis Logo
Lonely Planet Logo
Perplexity Logo
Quora Logo
Ricoh Logo
RO Logo
Robin AI Logo
SAP Logo
Scale Logo
SKT Logo
Slack Logo
Sourcegraph Logo
ZoomInfo Logo

Use cases for Claude

  • Operational efficiency

    Claude can extract relevant information from business emails and documents, categorize and summarize survey responses, and wrangle reams of text with high speed and accuracy.

  • Legal applications

    Contract drafting and negotiations require meticulous detail over extremely long, sensitive documents. Claude has a best-in-class 200K context window to help expedite traditionally manual legal reviews.

  • Code generation

    Claude models are constantly improving on coding, math, and reasoning. The latest model Claude 3 Opus scored 84.9% (up from 71.2%) on the Codex HumanEval, a Python coding test.