Model Square
DeepSeek

deepseek-v4-flash

DeepSeek·Dynamic billing

Official DeepSeek V4 Flash release with enhanced agentic capabilities and integrated DSpark speculative decoding

TPS
Average latency
Success rate

Pricing

Base Price

Input
$0.441176/ 1M
Output
$1.3235/ 1M
Cache Read$0.0147/ 1M
Dynamic billing
Prices vary by usage tier and request conditions
Tiered price table
peak
Input
$0.4412 / 1M
Output
$1.3235 / 1M
Cache Read
$0.0147 / 1M
off_peak
Input
$0.2206 / 1M
Output
$0.6618 / 1M
Cache Read
$0.0074 / 1M

Pricing by Group

default1x
TierInputOutputCache Read
peak
$0.441176
$1.3235
$0.0147
off_peak
$0.2206
$0.6618
$0.0074

Prices shown per 1M tokens

Model

ProviderDeepSeek
TypeDynamic billing
Groups
default
Endpoints
openai
Tags
ReasoningToolsOpenWeights1M
deepseek-v4-flash