DeepSeek API pricing, in your timezone

DeepSeek published the new rate card on 2026-08-13. Peak output on DeepSeek V4-Pro goes from $0.87 to $3.96 per million tokens — 4.6× the old rate.

The company has not published a cost breakdown, so any explanation beyond the announcement itself is speculation. What the numbers show is a move away from the aggressive undercutting that defined DeepSeek’s earlier pricing, toward rates closer to the rest of the market.

The new rates apply from 16 August 2026, 16:00 UTC. Requests billed before that instant use the old card; requests after it use the new one.

Because billing runs on UTC rather than your local clock, the switch lands mid-working-day in some regions. The countdown at the top of this site tracks it in your own 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.

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.

Input tokens are billed at two different rates. Tokens DeepSeek can serve from its prompt cache cost $0.044 per million at peak; tokens it has to process fresh cost $1.32 per million — a gap of 30×.

Caching keys on identical prefixes, so it rewards keeping the stable part of a prompt — system instructions, tool definitions, a file you are iterating on — byte-identical at the front, with the varying part at the end. Reordering that prefix between calls throws the cache away.

Coding assistants and long agent loops re-send the same context repeatedly and commonly sit above 90% cache hits. That is why the calculator on this site defaults to 80% rather than assuming every token is billed at full price.

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.

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.

The rate card covers API usage, billed per token against your API account balance.

Consumer access through the DeepSeek app and website is a separate product on separate terms, and is not what the prices on this site describe.

Balance is held in currency, not in tokens. Credit already on your account stays valid, but from the change-over each request draws down that balance at the new per-token rates, so a given balance buys proportionally fewer tokens than it did before.

Check the official pricing page for the terms that actually govern your account: https://api-docs.deepseek.com/quick_start/pricing/