← LLM API prices

DeepSeek V4.1 Flash

DeepSeek

DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...

In
$0.30
Out
$1.20
Blended
$0.52
Where to buy this model
17 access routes
RouteInOutBlendedvs
Relace cheapestfp4$0.15$0.60$0.26-
DeepInfra fp8$0.20$0.60$0.30+15%
Fireworks $0.22$0.66$0.33+27%
Morph fp8$0.21$0.84$0.37+42%
Reka fp4$0.29$1.16$0.51+96%
Alibaba $0.30$1.20$0.52+100%
Together $0.30$1.20$0.52+100%
SiliconFlow fp8$0.30$1.20$0.52+100%
Modal $0.30$1.20$0.52+100%
Wafer $0.30$1.20$0.52+100%
BaseTen fp8$0.30$1.20$0.52+100%
Parasail fp8$0.30$1.20$0.52+100%
GMICloud fp8$0.30$1.20$0.52+100%
Novita fp8$0.30$1.20$0.52+100%
DeepSeek $0.30$1.20$0.52+100%
Phala $0.34$1.38$0.60+131%
Venice fp8$0.38$1.50$0.66+154%

official API · cloud platform · inference host · USD per 1M tokens. Same model, different providers - price and quantization vary.

Ctx
1.048576M
max output
384K
cache read /1M
$0.01
Family
deepseek
Modalities
text · image
Features
json · reasoning · tools

Live pricing, updated 2026-09-15.

Concepts on this page