Skip to main content
GlobalFeaturedtop-picksglobal
Anthropic logo

Anthropic

Official Claude API access for Claude Fable 5.1, Opus 5, Sonnet 5, and Haiku 4.5, spanning long-horizon agents, coding, balanced production workloads, and high-throughput tasks.

getModelKey links official API key pages only. We do not sell, generate, or broker API keys.

Overview

Anthropic's current Claude lineup includes Fable 5.1 for the most demanding long-horizon agent work, Opus 5 for complex production coding and reasoning, Sonnet 5 for a balance of intelligence and cost, and Haiku 4.5 for fast, high-volume workloads.

As verified on September 3, 2026, standard input/output prices per million tokens are $10/$50 for Fable 5.1, $5/$25 for Opus 5, $3/$15 for Sonnet 5, and $1/$5 for Haiku 4.5. Fable 5.1 cache reads cost $0.25/MTok.

How to Get an API Key

  1. Sign up or log in at console.anthropic.com.
  2. Open Settings → API Keys and create a key.
  3. Copy the secret immediately and store it in a password manager or environment variable.
  4. Call the Messages API with the x-api-key header or an official Anthropic SDK.

Current Claude model family

Capability, composed in four voices

Use Fable for maximum capability, Opus for complex agentic work, Sonnet for balance, or Haiku for speed and cost.

USD per 1M tokens · Claude API

Claude Fable 5

Anthropic’s most capable broadly released model for long-running agents.

Model ID
claude-fable-5
Input
$10
5m cache write
$12.50
1h cache write
$20
Cache hit
$1
Output
$50
Context
1M
Max output
128K
Reliable knowledge
Jan 2026
Thinking
Adaptive, always on
Latency
Slower

Claude Opus 5

Complex agentic coding and enterprise work with adaptive thinking.

Model ID
claude-opus-5
Input
$5
5m cache write
$6.25
1h cache write
$10
Cache hit
$0.50
Output
$25
Context
1M
Max output
128K
Reliable knowledge
May 2026
Thinking
Adaptive
Latency
Moderate

Claude Sonnet 5

The strongest balance of speed and intelligence, with introductory pricing.

Model ID
claude-sonnet-5
Input
$2
5m cache write
$2.50
1h cache write
$4
Cache hit
$0.20
Output
$10
Context
1M
Max output
128K
Reliable knowledge
Jan 2026
Thinking
Adaptive
Latency
Fast

Claude Haiku 4.5

Anthropic’s fastest model with near-frontier intelligence.

Model ID
claude-haiku-4-5-20251001
Input
$1
5m cache write
$1.25
1h cache write
$2
Cache hit
$0.10
Output
$5
Context
200K
Max output
64K
Reliable knowledge
Feb 2025
Thinking
Extended thinking
Latency
Fastest
Sonnet 5 introductory input/output pricing is $2/$10 through August 31, 2026. Standard $3/$15 pricing begins September 1, 2026.
Official Anthropic snapshot checked August 10, 2026.Explore the complete Claude pricing guide →

Common Models

Claude Fable 5.1Claude Opus 5Claude Sonnet 5Claude Haiku 4.5

FAQ

Which current Claude model should I choose?

Start most demanding production workloads with Opus 5. Use Fable 5.1 when Opus at high effort falls short on difficult long-horizon work, Sonnet 5 for balanced production use, or Haiku 4.5 for latency-sensitive high-volume tasks.

What is the current Claude Sonnet 5 price?

Standard Sonnet 5 pricing is $3 input and $15 output per million tokens. Its $2/$10 introductory price ended August 31, 2026.

Can prompt caching reduce Claude API costs?

Yes. Fable 5.1 cache reads cost $0.25 per million tokens, while its 5-minute and 1-hour cache writes cost $12.50 and $20. Cache rates differ by model, so check the current pricing table before estimating savings.

Tags

reasoningcodinglong-context

Quick Info

Region
Global
Status
active
Last Verified
2026-09-03