DeepSeek V4 Flash 0423 — provider prices

17 providers offer this model. Cheapest input is $0.050, priciest is $0.600 — a 12.0x spread.

Providers
17
Cheapest
$0.050
Priciest
$0.600
Spread
12.0x

All quotes, cheapest first

ProviderInputOutput ContextQuantization
Relace$0.050$0.4501049KFP4 — lowest precision, expect quality loss
StreamLake$0.083$0.1651024KFP8 — good balance
DeepInfra$0.090$0.1801049KFP8 — good balance
GMICloud$0.091$0.1821049KFP8 — good balance
Venice$0.097$0.1931000Knot disclosed
DigitalOcean$0.098$0.1961049Knot disclosed
Fireworks$0.105$0.2101049Knot disclosed
SiliconFlow$0.130$0.2801049KFP8 — good balance
Crusoe$0.140$0.2801049KFP8 — good balance
Baidu$0.140$0.2801049KFP8 — good balance
AtlasCloud$0.140$0.2801049KFP4 — lowest precision, expect quality loss
Parasail$0.140$0.2801049KFP8 — good balance
Alibaba$0.150$0.3001000KFP8 — good balance
Mancer 2$0.190$0.5001049KFP8 — good balance
Azure$0.210$0.5601049Knot disclosed
Cloudflare$0.440$1.32384Knot disclosed
Wafer$0.600$0.1701049Kmxfp4

USD per million tokens. Verified 2026-11-04. Source: OpenRouter public API.

Cheapest is Relace at $0.050 input. Check the quantization column before switching: undisclosed or FP4 quotes can produce noticeably worse output on reasoning and code tasks.

Related

All DeepSeek models · Price ranges · Cost calculator · Official pricing