DeepSeek off-peak hours
DeepSeek charges less for requests that arrive outside its busiest hours — 50% off, applied to input and output alike. The windows are fixed to the UTC clock, so the discount lands at a different local hour for every reader of this page.
Live billing status
DeepSeek bills on a UTC clock, so the discounted window lands at a different hour depending on where you are. Enable JavaScript to see the current state in your timezone.
--:--:--
until the rate changes
- Peak output
- $3.96/M
- Off-peak output
- $1.98/M
Today where you are
UTC UTC+00:00- Peak 01:00 – 04:00 UTC
- Peak 06:00 – 10:00 UTC
- Off-peak all remaining hours
17h of every day is discounted by 50%. Billing follows the UTC clock, not your local one — shifting a batch job by a couple of hours is often the cheapest optimisation available.
The windows, in UTC
Peak hours are the periods below. Everything outside them — 17h of each day — bills at the off-peak rate.
Peak
- 01:00 – 04:00 UTC
- 06:00 – 10:00 UTC
Output billed at $3.96 per million tokens.
Off-peak
All remaining hours · 17h/day
Output billed at $1.98 per million tokens.
Peak hours by city
Local start and end of each peak window. A +1 or −1 marks a window that crosses local midnight. Cities that observe daylight saving shift by an hour twice a year while the UTC windows stay put.
| City | Peak 01:00–04:00 UTC | Peak 06:00–10:00 UTC |
|---|---|---|
| Los Angeles America/Los_Angeles | 18:00 −1 – 21:00 −1 | 23:00 −1 – 03:00 |
| New York America/New_York | 21:00 −1 – 00:00 | 02:00 – 06:00 |
| São Paulo America/Sao_Paulo | 22:00 −1 – 01:00 | 03:00 – 07:00 |
| London Europe/London | 02:00 – 05:00 | 07:00 – 11:00 |
| Berlin Europe/Berlin | 03:00 – 06:00 | 08:00 – 12:00 |
| Dubai Asia/Dubai | 05:00 – 08:00 | 10:00 – 14:00 |
| Bengaluru Asia/Kolkata | 06:30 – 09:30 | 11:30 – 15:30 |
| Shanghai Asia/Shanghai | 09:00 – 12:00 | 14:00 – 18:00 |
| Tokyo Asia/Tokyo | 10:00 – 13:00 | 15:00 – 19:00 |
| Sydney Australia/Sydney | 11:00 – 14:00 | 16:00 – 20:00 |
Snapshot taken for 16 August 2026, 16:00 UTC. The live panel above always reflects your own browser’s timezone and current DST state.
Capturing the discount
Move batch work
Evaluation runs, bulk classification and nightly summarisation rarely care about latency. Scheduling them inside the off-peak window halves their bill with no code change.
Schedule in UTC
A cron expression written in local time will drift out of the discount window when daylight saving changes. Pin the schedule to UTC and it stays aligned year round.
Watch the boundary
Billing follows when the request reaches DeepSeek. A long job started just before the window closes can finish on the more expensive side of the line.
Off-peak questions
What are DeepSeek’s off-peak hours in my timezone?
DeepSeek defines its windows on the UTC clock. Peak hours run 01:00–04:00 UTC and 06:00–10:00 UTC; every other minute of the day is off-peak, which works out to 17h of discounted time per day.
Where you are decides how convenient that is. In the mainland United States the windows fall overnight and in the evening, leaving a continuous off-peak run of at least 15 hours that covers the whole 9-to-5 working day in every US timezone, year round.
The table on this site converts the windows to whatever timezone your browser reports.
How much cheaper is the DeepSeek off-peak rate?
Off-peak rates are 50% lower than peak, applied to input and output alike. On DeepSeek V4-Pro that is $1.98 per million output tokens instead of $3.96.
It is worth being clear about what the discount is measured against: it is half of the new peak rate, not a return to the old one. Off-peak output still costs 2.3× what the same tokens cost under the previous flat card.
The rate is decided by when the request reaches DeepSeek, not when your job was queued locally. Batch work that can tolerate delay is the clearest way to capture it.
What is the difference between V4-Pro and V4-Flash?
Both carry a 1M token context window and the same maximum output length. They differ on price and on how many requests you may run at once: DeepSeek V4-Pro allows 500 concurrent requests, DeepSeek V4-Flash allows 2,500.
DeepSeek V4-Flash output costs $1.32 per million at peak against $3.96 for DeepSeek V4-Pro. Both models moved to the new card on the same date and share the same peak windows.
Is DeepSeek still cheaper than Claude, OpenAI and Gemini after the increase?
For most workloads, yes — but by a smaller margin than before, and the gap narrows further against the cheaper tiers of each provider once you account for cache hits.
The right comparison depends on your own mix of input, output and cache rate, which is what the calculator is for: it applies your volumes to every rate card at once instead of comparing headline numbers that assume a workload you may not have.