Blog Model Guides
DeepSeek V4 Peak and Off-Peak Routing: Verify the Window First
Treat a peak or off-peak discount as a testable contract, not a hard-coded date table.
AIWave prices checked 2026-11-05 UTC against the live pricing response and active default-group routes. The old price and capacity examples on this URL were not current evidence. Check the live rate card before using a number in production.
| Exact model ID | Fresh input | Cache-hit input | Output |
|---|---|---|---|
deepseek-v4-flash | $0.30 | $0.006 | $1.20 |
deepseek-v4-pro | $1.32 | $0.044 | $3.96 |
What is verified here
The table below is the checked AIWave default-group text-token rate on 2026-11-05 UTC. It does not establish a current peak/off-peak program. An earlier draft embedded an August schedule and multiple conflicting price windows; those figures had no current, request-level proof and have been removed.
How to validate a time-based route
- Obtain the provider's original effective window, timezone, model IDs, and rate conditions.
- Record the actual UTC dispatch time and effective route for a small controlled request at each boundary.
- Compare the provider's charge with AIWave's settled request record; do not infer a discount from a price page alone.
- Only enable a timed policy after observed settlement matches the documented window, with an immediate off switch.
Until those checks pass, use the live base rate and route by measured task quality, latency, and availability. If a time window is later verified, publish its original source and effective date alongside any number.
For an executable first request, use the quickstart. Recheck the effective account group and request settlement before treating an estimate as a bill.