top of page
Search

Claude AI All Models Available: catalog, pricing, and enterprise use.

ree

Claude AI now offers a complete model lineup that ranges from fast and cost-efficient variants to frontier reasoning systems. The catalog is structured into three main families—Sonnet, Opus, and Haiku—each optimized for different workloads and budgets. In 2025, Anthropic introduced Sonnet 4.5 with enhanced coding and agent support, and Opus 4.1 as its new frontier reasoning model, keeping both within their existing pricing tiers. With model aliases, long-context support, and enterprise deployment options, Claude AI provides a comprehensive selection that can adapt to individuals, teams, and regulated industries.

·····

.....

How the Claude model catalog is organized.

The Claude lineup is positioned across three categories:

  • Sonnet models balance cost and intelligence, serving as the default choice for most users. The latest Sonnet 4.5 improves performance in agents, coding, and computer-use tasks, while retaining the same pricing as Sonnet 4.

  • Opus models represent Anthropic’s highest reasoning tier, designed for advanced research, analysis, and long-form tasks. The new Opus 4.1 advances reasoning depth while staying priced at the Opus tier.

  • Haiku models are built for speed and efficiency, with low per-token costs suited for high-throughput applications and latency-sensitive deployments.

Anthropic maintains model aliases (e.g., claude-sonnet-4-5) and snapshot IDs (e.g., claude-sonnet-4-5-20250929) to ensure both stability and freshness in production use. This system allows developers to choose between fixed versions and automatically updated models.

·····

.....

Details of each Claude model.

Family

Model

Positioning

Context window

Max output

Notes

Sonnet

Sonnet 4.5

General-purpose, best for agents and coding

200k (1M beta available)

64k tokens

Latest release, same price as Sonnet 4

Sonnet

Sonnet 4

Balanced, reliable for all-round tasks

200k (1M beta available)

64k tokens

Available to free users in the Claude app

Sonnet

Sonnet 3.7

Legacy with extended thinking

200k

64k tokens

Remains available for compatibility

Opus

Opus 4.1

Frontier reasoning, advanced research

200k

32k tokens

Successor to Opus 4, same price tier

Opus

Opus 4

Frontier model, earlier flagship

200k

32k tokens

Available across app and API

Haiku

Haiku 3.5

Fastest and most cost-efficient

200k

8,192 tokens

Best for scale and latency-sensitive workloads

Haiku

Haiku 3

Legacy fast model

200k

4,096 tokens

Older but still supported in some stacks

This catalog shows how Claude covers the full spectrum of needs, from large-context research to real-time applications.

·····

.....

Pricing of Claude models.

Anthropic publishes detailed per-million token pricing for all Claude models, with consistent rates across API and partner platforms.

Model

Input price

Output price

Cache (5 min / 1h)

Cache refresh

Opus 4.1

$15

$75

$18.75 / $30

$1.50

Opus 4

$15

$75

$18.75 / $30

$1.50

Sonnet 4.5

$3

$15

$3.75 / $6

$0.30

Sonnet 4

$3

$15

$3.75 / $6

$0.30

Sonnet 3.7

$3

$15

$3.75 / $6

$0.30

Haiku 3.5

$0.80

$4

$1.00 / $1.60

$0.08

Haiku 3

$0.25

$1.25

$0.30 / $0.50

$0.03

Sonnet 4.5 retains the Sonnet 4 price point, while Opus 4.1 remains aligned with Opus 4. This ensures continuity for developers and enterprises budgeting around existing workloads.

·····

.....

Where Claude models are available.

Claude models are deployed across multiple access paths:

  • Claude app (consumer): Sonnet 4 is available for free users; paid plans (Pro, Max, Team, Enterprise) include Sonnet and Opus variants, with Sonnet 4.5 rolling out broadly.

  • Anthropic API: All models (Sonnet 4.5/4, Opus 4.1/4, Haiku 3.5/3, Sonnet 3.7) are accessible with snapshot IDs and stable aliases.

  • AWS Bedrock: Carries Claude Sonnet and Opus variants, including the latest Sonnet 4.5.

  • Google Vertex AI: Hosts Claude models in Model Garden, with governance and FedRAMP compliance for enterprise use.

This distribution ensures that developers, businesses, and individuals can access Claude from whichever ecosystem they operate in.

·····

.....

Plans and subscriptions.

Claude’s subscription model is designed for individuals, teams, and enterprises:

Plan

Price

Features

Free

$0

Sonnet 4 with daily usage caps

Pro

$20/month or $200/year

Higher message caps, access to Sonnet and Opus

Max

$100–$200/month

Extended quotas, optional extra usage at API rates

Team Standard

$25/user/month (annual) or $30/user/month (monthly), 5-user minimum

Centralized billing, collaboration, admin tools

Team Premium

$150/user/month

Higher usage, Claude Code, intensive team workloads

Enterprise

Custom

Governance, SSO, audit controls, high quotas, API overages billed by tokens

This structure allows Anthropic to serve casual users, professionals, and global enterprises with predictable costs and clear governance controls.

·····

.....

Context window and output capacities.

Most Claude models support a 200,000-token context window, with Sonnet 4 and Sonnet 4.5 offering an extended 1 million-token beta option. Output ranges vary by model, with up to 64,000 tokens for Sonnet, 32,000 for Opus, and smaller outputs for Haiku variants. These extended limits allow Claude to handle book-length documents, full research reports, and complex codebases in a single session.

·····

.....

Practical recommendations for choosing a model.

  • Sonnet 4.5 is the best general choice for individuals and teams who need a balance of speed, intelligence, and cost efficiency, especially in workflows involving coding and agents.

  • Opus 4.1 should be reserved for the most demanding reasoning tasks, such as deep research or high-stakes enterprise use, where output quality outweighs cost.

  • Haiku 3.5 is ideal for high-volume workloads, real-time applications, and situations where latency and scale matter more than reasoning depth.

  • Enterprise users should deploy Claude through Bedrock, Vertex AI, or Anthropic’s own Enterprise offering to ensure compliance, quota scaling, and governance.

By aligning model choice with task complexity, budget, and compliance needs, individuals and organizations can optimize how they use Claude AI’s model suite across research, development, and production environments.

.....

FOLLOW US FOR MORE.

DATA STUDIOS

bottom of page