DeepSeek V4.1 Flash — provider prices

40 providers offer this model. Cheapest input is $0.033, priciest is $0.450 — a 13.6x spread.

Providers
40
Cheapest
$0.033
Priciest
$0.450
Spread
13.6x

All quotes, cheapest first

ProviderInputOutput ContextQuantization
Baidu$0.033$0.1321049KFP8 — good balance
OpenInference$0.035$0.4751049KFP4 — lowest precision, expect quality loss
Morph$0.035$0.3151049KFP8 — good balance
Relace$0.035$0.5001049KFP4 — lowest precision, expect quality loss
Wafer$0.045$0.5001049Kmxfp4
Sail Research$0.090$0.2501049KFP4 — lowest precision, expect quality loss
Sail Research$0.090$0.2501049KFP4 — lowest precision, expect quality loss
InferenceNet$0.103$0.6001040KFP8 — good balance
InferenceNet$0.103$0.6001040KFP8 — good balance
Reka$0.105$0.8401049Knot disclosed
DeepInfra$0.112$0.3361049KFP8 — good balance
CoreWeave$0.130$0.6501049KFP8 — good balance
AtlasCloud$0.135$0.5401049KFP8 — good balance
DeepInfra$0.140$0.4201049KFP8 — good balance
StreamLake$0.147$0.5881024KFP8 — good balance
Mancer 2$0.150$0.7501049KFP8 — good balance
PrimeIntellect$0.156$1.201049Knot disclosed
Alibaba$0.165$0.6601000Knot disclosed
GMICloud$0.180$0.7201049KFP8 — good balance
Wafer$0.190$0.6501049Kmxfp4
Novita$0.195$0.7801049KFP8 — good balance
DeepInfra$0.210$0.6301049KFP8 — good balance
Fireworks$0.210$0.8401049Knot disclosed
Phala$0.210$0.8401049Knot disclosed
DigitalOcean$0.225$0.9001049Knot disclosed
Makora$0.260$1.171049KFP8 — good balance
Crusoe$0.290$1.201049KFP8 — good balance
Relace$0.300$0.9001049KFP4 — lowest precision, expect quality loss
Google$0.300$1.201049Knot disclosed
Nebius$0.300$1.201049KFP8 — good balance
Cohere$0.300$1.201049Knot disclosed
BaseTen$0.300$1.201049Knot disclosed
Together$0.300$1.201049Knot disclosed
SiliconFlow$0.300$1.201049KFP8 — good balance
Modal$0.300$1.201049Knot disclosed
BaseTen$0.300$1.201049Knot disclosed
Parasail$0.300$1.201049KFP8 — good balance
Fireworks$0.300$1.201049Knot disclosed
Venice$0.300$1.201000KFP8 — good balance
Fireworks$0.450$1.801049Knot disclosed

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

Cheapest is Baidu at $0.033 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