Current Anthropic Claude API Pricing 2026: Opus 5.5, Fable 5.1, Sonnet 5.5
Current Claude API prices for Opus 5.5, Fable 5.1, Sonnet 5.5, Opus 5 and Haiku 4.5, including prompt caching, Batch and Fast mode.
Anthropic’s current Claude API lineup is led by Claude Fable 5.1 for the highest broadly available capability, Claude Opus 5.5 as the default for most demanding work, Claude Sonnet 5.5 for balanced production traffic, and Claude Haiku 4.5 for speed and cost.
Claude API Prices Per 1M Tokens
| Model | Input | Cache read | 5m cache write | 1h cache write | Output | Context |
|---|---|---|---|---|---|---|
| Claude Fable 5.1 | $10.00 | $0.25 | $12.50 | $20.00 | $50.00 | 1M |
| Claude Opus 5.5 | $4.00 | $0.20 | $5.00 | $8.00 | $20.00 | 1M |
| Claude Sonnet 5.5 | $2.00 | $0.20 | $2.50 | $4.00 | $10.00 | 1M |
| Claude Sonnet 5 | $2.00 | $0.20 | $2.50 | $4.00 | $10.00 | 1M |
| Claude Fable 5 | $10.00 | $1.00 | $12.50 | $20.00 | $50.00 | 1M |
| Claude Opus 5 | $5.00 | $0.50 | $6.25 | $10.00 | $25.00 | 1M |
| Claude Opus 4.8 | $5.00 | $0.50 | $6.25 | $10.00 | $25.00 | 1M |
| Claude Haiku 4.5 | $1.00 | $0.10 | $1.25 | $2.00 | $5.00 | 200K |
Anthropic canceled Sonnet 5’s planned September increase: its launch rate of $2/$10 is now permanent. Fable 5.1 also cuts cache reads to 0.025 times base input, one quarter of Fable 5’s cache-read cost.
Sonnet 5.5 launched on September 28, 2026 at the same $2/$10 standard rate as Sonnet 5, with $0.20/M cache reads and $2.50/M five-minute cache writes across the full 1M context window.
Batch and Fast Mode
The Message Batches API discounts input and output by 50%:
| Model | Batch input | Batch output |
|---|---|---|
| Fable 5.1 | $5.00 | $25.00 |
| Opus 5.5 | $2.00 | $10.00 |
| Sonnet 5.5 | $1.00 | $5.00 |
| Haiku 4.5 | $0.50 | $2.50 |
Opus 5.5 Fast mode costs $8/M input and $40/M output for up to roughly 2.5x faster generation. US-only inference and eligible partner regional endpoints add a 10% premium.
Monthly Cost Example
For 2M uncached input plus 500K output tokens:
| Model | Standard cost |
|---|---|
| Claude Haiku 4.5 | $4.50 |
| Claude Sonnet 5.5 | $9.00 |
| Claude Opus 5.5 | $18.00 |
| Claude Opus 5 | $22.50 |
| Claude Fable 5.1 | $45.00 |
Opus 5.5 is 20% cheaper than Opus 5 for this cache-free token mix. Its advantage grows in agent loops because cache reads cost 60% less than Opus 5.
Which Claude Model Should You Use?
- Fable 5.1: the hardest long-horizon agents, research, coding, and document work after Opus 5.5 falls short in your evals.
- Opus 5.5: default premium model for long-running coding agents and professional knowledge work.
- Sonnet 5.5: first production baseline for coding, tool use, assistants, and balanced traffic.
- Haiku 4.5: extraction, classification, routing, short answers, and latency-sensitive work.
Read the Opus 5.5 value analysis for comparisons with GPT-6, or use the AI Model Pricing Calculator for your own traffic.
Official Sources
Related Posts
Is Claude Opus 5.5 Worth It? Pricing and Value vs GPT-6
2026-09-25
Is Claude Opus 4.8 Worth Upgrading To? Capability, Cost, and the Fable 5 Problem
2026-06-14
AI Coding Agent Cost Comparison 2026: GPT-6, Opus 5.5, DeepSeek, and Kimi
2026-05-07