Model Square
DeepSeek

grok-4.6

DeepSeek·Usage-based billing

grok-4.6

TPS
Average latency
Success rate

Pricing

Base Price

Input
$2/ 1M
Output
$6/ 1M
Cache Read$0.5/ 1M
Cache Write$2.5/ 1M
Cache Write (1h)$4/ 1M
Image In$2/ 1M
Audio In$2/ 1M
Audio Out$2/ 1M

Model

ProviderDeepSeek
TypeUsage-based billing
Groups
研发
Endpoints
openai
grok-4.6