Context window
1,000,000 tokens
Checked 19 Oct 2026https://api-docs.deepseek.com/quick_start/pricing
LLM · DeepSeek
DeepSeek V4 Pro
DeepSeek's API pricing page lists deepseek-v4-pro (DeepSeek-V4-Pro-0813) with a 1M context and 384K maximum output. The stored rate is the peak cache-miss price, $1.32 input and $3.96 output per 1M tokens. Off-peak is half, $0.66 and $1.98. Peak hours are Monday-Friday 01:00-04:00 and 06:00-10:00 UTC, excluding Chinese public holidays. The page does not name a license or say whether weights are downloadable.
Checked 19 Oct 2026https://api-docs.deepseek.com/quick_start/pricing
Input price
$1.32
Checked 19 Oct 2026https://api-docs.deepseek.com/quick_start/pricing
Output price
$3.96
Checked 19 Oct 2026https://api-docs.deepseek.com/quick_start/pricing
API deployment
Yes
Checked 19 Oct 2026https://api-docs.deepseek.com/quick_start/pricing
Local / self-hosted
Not published
Checked 19 Oct 2026https://api-docs.deepseek.com/quick_start/pricing
VPC / cloud native
Not published
Checked 19 Oct 2026https://api-docs.deepseek.com/quick_start/pricing
License
Not published
Checked 19 Oct 2026https://api-docs.deepseek.com/quick_start/pricing
Data retention
Not published
Checked 19 Oct 2026https://api-docs.deepseek.com/quick_start/pricing