DeepSeek
A family of open language models from DeepSeek
A family of large language and reasoning models developed by DeepSeek, including named generations and variants such as DeepSeek-V3 and DeepSeek-R1.
Pricing
Official site · Nov 8, 2026, 2:07 PM
Input / 1M
$0.15
Output / 1M
$0.60
Cached input / 1M
$0.003
Representative deepseek-flash rates for off-peak usage: $0.15/1M input cache-miss tokens, $0.003/1M input cache-hit tokens, and $0.60/1M output tokens. Peak rates are $0.30, $0.006, and $1.20 respectively. The same page lists different rates for deepseek-v4-pro.
Official DeepSeek API documentation lists per-1M-token prices for current model endpoints, including deepseek-flash. This finding uses the deepseek-flash rate as the representative DeepSeek text API price; the API has multiple models and peak/off-peak rates.
Model Intelligence
Arena ranking
5
Benchmarkable
No
Model level
family
Intelligence Index
30.4
Coding Index
59.4
GPQA
0.89
HLE
0.38
SciCode
0.51
IFBench
0.76
LCR
0.75
TerminalBench Hard
0.46
TAU2
0.96
Recent stories
1 linked story