Every Anthropic model
Each model on its default cache setting. Select a model for its worked examples, break-even rule and caching rules.
| Model | Input | 5 min write | 1 hour write | Cache read | Output | Min. cacheable | Coding agent | Without | Share saved |
|---|---|---|---|---|---|---|---|---|---|
| Claude Fable 5.1 | $10.00 | $12.50 | $20.00 | $0.25 | $50.00 | 512 | $8.02 | $38.48 | 79% |
| Claude Fable 5 | $10.00 | $12.50 | $20.00 | $1.00 | $50.00 | 512 | $10.46 | $38.48 | 73% |
| Claude Opus 5.5 | $4.00 | $5.00 | $8.00 | $0.20 | $20.00 | 512 | $3.53 | $15.39 | 77% |
| Claude Opus 5 | $5.00 | $6.25 | $10.00 | $0.50 | $25.00 | 512 | $5.23 | $19.24 | 73% |
| Claude Opus 4.8 | $5.00 | $6.25 | $10.00 | $0.50 | $25.00 | 1,024 | $5.23 | $19.24 | 73% |
| Claude Opus 4.7 | $5.00 | $6.25 | $10.00 | $0.50 | $25.00 | 2,048 | $5.23 | $19.24 | 73% |
| Claude Opus 4.6 | $5.00 | $6.25 | $10.00 | $0.50 | $25.00 | 4,096 | $5.23 | $19.24 | 73% |
| Claude Sonnet 5.5 | $2.00 | $2.50 | $4.00 | $0.10 | $10.00 | 512 | $1.77 | $7.70 | 77% |
| Claude Sonnet 5 | $2.00 | $2.50 | $4.00 | $0.20 | $10.00 | 1,024 | $2.09 | $7.70 | 73% |
| Claude Sonnet 4.6 | $3.00 | $3.75 | $6.00 | $0.30 | $15.00 | 1,024 | $3.14 | $11.54 | 73% |
| Claude Haiku 5.5 | $0.10 | $0.125 | $0.20 | $0.01 | $0.50 | 512 | $0.37 | $1.36 | 73% |
| Claude Haiku 4.5 | $1.00 | $1.25 | $2.00 | $0.10 | $5.00 | 4,096 | $1.05 | $3.85 | 73% |
Coding agent: 40 turns, 20K tokens of starting context, 3K in and 600 out per turn, 1.5 minutes apart, three long breaks of 8 minutes.
How Anthropic caching works
- You mark cache breakpoints, or turn on automatic caching, which moves the breakpoint to the last cacheable block as the conversation grows.
- Two lifetimes: 5 minutes, the default, and 1 hour. Writes cost more than plain input. Reads cost a small fraction of it and refresh the lifetime for free.
- Each model has a minimum cacheable prompt of 512, 1,024, 2,048 or 4,096 tokens. Shorter prompts are processed without caching.
- Haiku 5.5 is priced by whole prompt length, with higher prices above 100K tokens. Claude 4.6 and later have no long-context premium otherwise.
Sources: pricing page and caching guide. Each model page lists the rules that apply to it.
Other providers
OpenAI Google All 32 models compared
Try any of these sessions with your own numbers in the simulator. It runs in your browser and shows where the cache goes cold.