Anthropic Model Pricing
October 12, 2026: claude-haiku-5-5 is the current low-cost Haiku: $0.10 input / $0.50 output per million tokens through 100K prompt tokens; above that, the full request uses $0.50 / $2.50. Prompt length includes cache reads and writes. Five-minute cache writes cost $0.125 or $0.625, one-hour writes $0.20 or $1, and reads $0.01 or $0.05 in the respective brackets. Batch input/output is half Standard. Sonnet 5.5 cache reads now cost $0.10/M, not $0.20. First-party pricing. Haiku 4.5 rows below are legacy references, not the current-generation recommendation.
All prices per 1 million tokens. Data verified Oct 14, 2026.
| Model | Input (per 1M) | Output (per 1M) | Context | Tier |
|---|---|---|---|---|
| Claude Opus 5 NEW | $5.00 | $25.00 | 1M | Active |
| Claude Fable 5 NEW | $10.00 | $50.00 | 1M | Premium |
| Claude Mythos 5 NEW Restricted | $10.00 | $50.00 | 1M | Premium |
| Claude Opus 4.8 | $5.00 | $25.00 | 1M | Premium |
| Claude Opus 4.7 | $5.00 | $25.00 | 1M | Premium |
| Claude Opus 4.6 | $5.00 | $25.00 | 1M | Premium |
| Claude Opus 4.1 Retires Aug 5 | $15.00 | $75.00 | 200K | Premium |
| Claude Sonnet 5.5 CURRENT | $2.00 | $10.00 | 1M | Mid |
| Claude Sonnet 5 Previous | $2.00 | $10.00 | 1M | Mid |
| Claude Sonnet 4.6 | $3.00 | $15.00 | 1M | Mid |
| Claude Sonnet 4 Retired | $3.00 | $15.00 | 200K | Mid |
| Claude Haiku 5.5 NEW | $0.10 | $0.50 | 1M | Budget |
| Claude Haiku 4.5 | $1.00 | $5.00 | 200K | Budget |
Claude Haiku 5.5 pricing is tiered by prompt length: $0.10 input / $0.50 output for prompts up to 100,000 tokens, and $0.50 input / $2.50 output for prompts over 100,000 tokens. The table shows the base (≤100K) rates. Released October 7, 2026.
Which Claude Model Should You Use?
Choose the right model based on your workload, quality requirements, and budget.
Complex Reasoning & Analysis
Multi-step reasoning, research synthesis, legal analysis, and tasks requiring the highest accuracy. Best-in-class performance on graduate-level benchmarks.
Code Generation & Agentic Tasks
Writing, debugging, and reviewing code. Building multi-step workflows and tool-use agents. Strong balance of capability and speed.
High-Volume Chat & Classification
Customer support chatbots, content moderation, sentiment analysis, and classification tasks. Excellent quality at a fraction of the cost.
Content Generation at Scale
Summarization, translation, and content creation where throughput matters. 1M context handles large documents with ease.
Enterprise Document Processing
Analyzing lengthy contracts, research papers, or codebases. The 1M context window processes entire documents in a single call.
Startup MVPs & Prototyping
Rapid prototyping and MVP development where cost efficiency is critical. Start with Haiku for most tasks, upgrade to Sonnet when needed.
Calculate Your Anthropic Costs
Estimate your monthly spend across all Claude models.
Monthly Cost Estimate
Anthropic vs Competitors
How Claude models stack up against equivalent models from other providers.
| Tier | Anthropic | Input / Output | OpenAI | Input / Output | Input / Output | |
|---|---|---|---|---|---|---|
| Budget | Claude Haiku 5.5 | $0.10 / $0.50 | GPT-5.4 nano | $0.20 / $1.25 | Gemini 2.5 Flash-Lite | $0.10 / $0.40 |
| Mid | Claude Sonnet 5.5 | $2.00 / $10.00 | GPT-5.4 | $2.50 / $15.00 | Gemini 3.1 Pro | $2.00 / $12.00 |
| Premium | Claude Opus 5 | $5.00 / $25.00 | GPT-5.5 | $5.00 / $30.00 | Gemini 3.1 Pro | $2.00 / $12.00 |
Note: Competitor pricing is for comparable capability tiers. Actual performance may vary by task. Google Gemini pricing reflects paid tier rates.
See how Claude models stack up against OpenAI, Google, and more
Deep-Dive Comparisons
Read our detailed head-to-head analyses of Claude models vs the competition.
Gemini 3.5 Flash vs Claude Haiku 4.5
Budget AI showdown: Haiku 4.5 ($1/$5) vs Gemini 3.5 Flash ($1.50/$9). Haiku is 33% cheaper input, 44% cheaper output.
Command A vs Claude Sonnet 5.5
Command A ($2.50/$10) vs Claude Sonnet 5.5 ($2/$10). Compare workload fit, context and total task cost rather than token rates alone.
Fable 5 vs GPT-5.5
The premium AI API showdown: Fable 5 ($10/$50) vs GPT-5.5 ($5/$30). Which flagship justifies its price?
Claude Opus 4.8 vs Gemini 3.1 Pro
Two 1M-context flagship models — Gemini is 52% cheaper on output. Full pricing and use-case comparison.
Claude Sonnet 4.6 vs GPT-5
GPT-5 costs 58% less but Sonnet 4.6 offers 3.7x more context. Full pricing and performance comparison.
Calculate Your Anthropic Costs
Enter your usage and see exactly what each Claude model would cost you — no signup required.
Open the Cost Calculator