deepseek
DeepSeek: DeepSeek V4 Flash 0731
deepseek/deepseek-v4-flash-0731
For $1, you can send approximately:
~220messages
How do we get this number?
One message = ~7,000 input tokens + ~7,000 output tokens
Input cost per message7,000 x $0.0080/M = $0.000056
Output cost per message7,000 x $0.64/M = $0.004480
Total cost per message$0.004536
Messages for $1220.46
Context window
1.0M
tokens
Max response
944k
tokens
Input price
$0.0080
per million tokens
Output price
$0.64
per million tokens
Modalities
Input:textOutput:text
Description
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....