Models & rates
Every Claude and ChatGPT model, one rate table.
24 models, 17 from Claude and 7 from ChatGPT. Every model in a tier bills at the same public rate, so you can pick by capability instead of by price list.
- Frontier
- $4.20 in · $28.00 out
- Balanced
- $1.30 in · $6.00 out
- Speed
- $1.30 in · $6.00 out
USD per 1M tokens
Model explorer
Showing 24 of 24 models
| Model | Tier | Context | Input / 1M | 5m cache write | 1h cache write | Cached input / 1M | Output / 1M |
|---|---|---|---|---|---|---|---|
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Balanced | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Balanced | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Balanced | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Speed | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Balanced | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Frontier | 262K | $4.20 | $4.20 | $4.20 | $2.45 | $28.00 | |
| Balanced | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Balanced | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Speed | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Speed | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Balanced | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 | |
| Speed | 131K | $1.30 | $1.30 | $1.30 | $1.30 | $6.00 |
Rates are in USD per 1M tokens, and each request is billed at the rate of the model that served it. Cache writes (5-minute and 1-hour) are billed at the input rate; reads from the prompt cache are billed at the cached-input rate. Relative speed is a guide based on tier, not a latency guarantee.