DeepSeek
Cheapest model in the catalog.
Served by DeepSeek-V4.1-Flash-1017. Peak pricing applies 01:00–04:00 and 06:00–10:00 UTC Mon–Fri; off-peak is half price. Cached input $0.003/M off-peak.
Latency
0.8s
Context
1M
Input
$0.15/M
Output
$0.6/M
{
"model": "deepseek-v4-flash",
"messages": [
{
"role": "user",
"content": "Hello, world."
}
]
}