SV Premium

LLM APIs

DeepSeek

DeepSeek's first-party API serving its open-weight (MIT) DeepSeek V4 Pro and V4 Flash models, with peak and off-peak pricing.

You can halve the listed rates by running outside the weekday peak windows of 01:00–04:00 and 06:00–10:00 UTC, or on weekends and Chinese public holidays. V4 Flash cached input costs $0.006 per million tokens. The first-party API processes data in China, with no EU residency or GDPR DPA. Other hosts also serve these MIT-licensed models; check their regions separately.

Headquarters
🇨🇳 China
Parent jurisdiction
🇨🇳 China (parent: High-Flyer)
EU self-serve plan
No / not documented
GDPR DPA
No / not documented
Billing
payg
Open source
No
Website
www.deepseek.com
Source last checked
2026-10-06
Pricing evidence
Published rates

Pricing

Calculated from the listed vendor rates; exclusions still apply. How to read these estimates.

ModelInput / 1M tokensOutput / 1M tokensCached / 1M tokensContextDetails
DeepSeek V4 Pro$1.32$3.96$0.0441000kfrontieropen weights
DeepSeek V4 Flash$0.3$1.2$0.0061000kmidopen weights

Prices shown are peak-hour rates (01:00-04:00 and 06:00-10:00 UTC, Mon-Fri). Off-peak hours, weekends and Chinese public holidays are billed at 50% of these rates. Data is processed in China. The model weights are MIT-licensed on Hugging Face, so the same models are also served by third-party hosts.

Managed alternatives

At 500M input tokens / month, 50M output tokens / month. Adjust usage and requirements.

DeepSeek · DeepSeek V4 Flash: $2,520/yr estimated.

The price difference leaves out migration costs, operating costs and any features you’d need to replace. We compare the selected model or the cheapest listed model in the chosen capability tier. Sharing a tier does not mean equal quality.

Found a wrong rate or plan restriction? Report a correction with a current source.