anthropicclaudepricingcomparison

Claude API Pricing 2026: Every Model, Every Tier Explained

Complete guide to Anthropic Claude API pricing — Opus, Sonnet, and Haiku tiers, prompt caching discounts, and how Claude compares to GPT-4o at scale.

TTokenCost Editorial·LLM Cost Research·Updated 2026-04-286 min read

Anthropic's Claude family covers three tiers — Opus (frontier), Sonnet (balanced), and Haiku (fast/cheap) — with distinct pricing at each level. All Claude 4 and Claude 3.5 models support prompt caching, which can reduce input costs by up to 90%. This guide covers every Claude model, their pricing, and how they compare to OpenAI.

All Claude Models — Pricing Table

Model	Input /1M	Cached /1M	Output /1M	Context
Claude Fable 5	$10	$1	$50	1M
Claude Opus 4.8	$5	$0.5	$25	1M
Claude Opus 4.7	$5	$0.5	$25	1M
Claude Sonnet 4.6	$3	$0.3	$15	1M
Claude Opus 4.6	$5	—	$25	1M
Claude Opus 4.5	$15	$1.5	$75	200k
Claude Haiku 4.5	$1	$0.1	$5	200k
Claude 3.5 Sonnet	$3	$0.3	$15	200k
Claude 3.5 Haiku	$0.8	$0.08	$4	200k
Claude 3 Opus	$15	$1.5	$75	200k

Claude 4 vs Claude 3.5: What Changed?

Claude 4 (Opus 4.7, Sonnet 4.6, Opus 4.6, Opus 4.5, Haiku 4.5) represents Anthropic's latest generation with improved reasoning, longer context handling, and better instruction following. All Claude 4 models support extended thinking (similar to OpenAI o3) for complex reasoning tasks. Claude 3.5 Sonnet and 3.5 Haiku remain excellent for cost-sensitive workloads and have proven reliability in production.

Claude Opus vs Sonnet vs Haiku: Cost Comparison

At 10,000 requests/day (2,000 input + 1,000 output tokens):

Claude Opus 4.7

$10500/mo

$5 in · $25 out /1M

Claude Sonnet 4.6

$6300/mo

$3 in · $15 out /1M

Claude Haiku 4.5

$2100/mo

$1 in · $5 out /1M

When to Use Each Tier

Claude Haiku 4.5

High-volume, cost-sensitive tasks

Customer support, classification, simple Q&A, chatbots

Claude Sonnet 4.6

Most production use cases

Complex reasoning, coding, analysis, content generation — best balance

Claude Opus 4.7

Maximum capability tasks

Autonomous agents, complex coding, research, frontier reasoning

Claude vs GPT-4o: Price Comparison

Model	Input /1M	Output /1M	Caching
Claude Sonnet 4.6	$3	$15	$0.3/1M
Claude Haiku 4.5	$1	$5	$0.1/1M
GPT-4o	$2.5	$10	$1.25/1M
GPT-4.1	$2	$8	$1/1M

Prompt Caching: Claude's Cost Advantage

All Claude models support prompt caching — and Anthropic offers the most aggressive caching discounts in the industry. If your prompts contain a large, repeated prefix (system prompt, few-shot examples, RAG context), caching can reduce your effective input cost by 50–90%.

Example: Claude Haiku 4.5 with 80% cache hit rate

Without caching

$1/1M input

Effective rate (80% cache)

$0.280/1M effective

View full Claude pricing: Anthropic API pricing → | Claude Sonnet vs GPT-4o →

Cheapest LLM API in 2026: Full Price Comparison

We compared 26 LLM models across 8 providers to find the cheapest API for every use case — from bulk processing to complex reasoning.

8 min read

GPT vs Claude vs Gemini: Pricing & Performance in 2026

A detailed comparison of OpenAI, Anthropic, and Google's pricing models, context windows, and value for different workloads.

7 min read