DeepSeekChat

DeepSeek V4 Flash

$0.049565% offper 1M tokens$0.1465% off

An efficiency-tuned mixture-of-experts model from DeepSeek, with 13B of 284B parameters active. It is built for fast inference.

1.05MContext
$0.0495Input / 1M
$0.099Output / 1M
65%Saved vs retail
client.chat.completions.create(    model="deepseek-v4-flash",    messages=[        {"role": "user", "content": "Hello"},    ],)

Cost per 1M input tokens

Studios$0.0495
Retail$0.14

Also new