The most notable change in V4 is that million-word-long context has become standard across all official services, with deep optimization for various agent tools.
Different token groups have different prices. Unit: Million tokens (M)
| Token Group | Group Ratio | Input Price | Output Price | Reasoning output | Cache Read |
|---|---|---|---|---|---|
| deepseek Deepseek | 0.5x | $2.25/M | $6.75/M | $6.75/M | $0.075/M |
Billed by the time window the request starts in; requests outside every window use the other-hours price. Times are in UTC
| Window | Applies | Input Price | Output Price | Reasoning output | Cache Read |
|---|---|---|---|---|---|
| Peak | Mon–Fri 01:00-04:00, 06:00-10:00 | $4.5/M | $13.5/M | $13.5/M | $0.15/M |
| Off-peak | Other hours | $2.25/M | $6.75/M | $6.75/M | $0.075/M |