Skip to content

DeepSeek V4.1 Flash

deepseek/deepseek-v4.1-flash

context
1.05M
modalities
text, image → text
capabilities
Tool use, Reasoning, Vision, Structured output, Long context, Low cost

Overview

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...

Providers

Pricing

  • Alibaba

    input / 1M
    $0.09
    output / 1M
    $0.36
  • AtlasCloud

    input / 1M
    $0.135
    output / 1M
    $0.54
  • Baidu

    input / 1M
    $0.0507
    output / 1M
    $0.2028
  • BaseTen

    baseten

    input / 1M
    $0.30
    output / 1M
    $1.20
  • BaseTen

    baseten/us

    input / 1M
    $0.30
    output / 1M
    $1.20
  • Cloudflare

    input / 1M
    $0.10
    output / 1M
    $0.90
  • Cohere

    input / 1M
    $0.195
    output / 1M
    $0.66
  • CoreWeave

    input / 1M
    $0.13
    output / 1M
    $0.65
  • Crusoe

    input / 1M
    $0.29
    output / 1M
    $1.20
  • DeepInfra

    deepinfra/fp8 · fp8

    input / 1M
    $0.14
    output / 1M
    $0.42
  • DeepInfra

    deepinfra/fp8/flex · fp8

    input / 1M
    $0.112
    output / 1M
    $0.336
  • DeepInfra

    deepinfra/fp8/priority · fp8

    input / 1M
    $0.21
    output / 1M
    $0.63
  • DigitalOcean

    input / 1M
    $0.195
    output / 1M
    $0.78
  • Fireworks

    fireworks

    input / 1M
    $0.30
    output / 1M
    $1.20
  • Fireworks

    fireworks/spot

    input / 1M
    $0.15
    output / 1M
    $0.60
  • Fireworks

    fireworks/us

    input / 1M
    $0.45
    output / 1M
    $1.80
  • GMICloud

    input / 1M
    $0.18
    output / 1M
    $0.72
  • Google

    google-vertex/global

    input / 1M
    $0.30
    output / 1M
    $1.20
  • InferenceNet

    inference-net/fp8 · fp8

    input / 1M
    $0.103
    output / 1M
    $0.60
  • InferenceNet

    inference-net/us · fp8

    input / 1M
    $0.103
    output / 1M
    $0.60
  • Ionstream

    input / 1M
    $0.20
    output / 1M
    $0.70
  • Makora

    input / 1M
    $0.26
    output / 1M
    $1.17
  • Mancer 2

    input / 1M
    $0.2125
    output / 1M
    $0.75
  • Modal

    input / 1M
    $0.30
    output / 1M
    $1.20
  • Nebius

    input / 1M
    $0.30
    output / 1M
    $1.20
  • Novita

    input / 1M
    $0.15
    output / 1M
    $0.60
  • OpenInference

    input / 1M
    $0.06
    output / 1M
    $0.80
  • Parasail

    parasail/fast · fp8

    input / 1M
    $0.45
    output / 1M
    $1.80
  • Parasail

    parasail/fp8 · fp8

    input / 1M
    $0.30
    output / 1M
    $1.20
  • Reka

    input / 1M
    $0.12
    output / 1M
    $0.64
  • Relace

    input / 1M
    $0.12
    output / 1M
    $0.50
  • River

    input / 1M
    $0.12
    output / 1M
    $0.48
  • SiliconFlow

    input / 1M
    $0.30
    output / 1M
    $1.20
  • StreamLake

    input / 1M
    $0.147
    output / 1M
    $0.588
  • Together

    input / 1M
    $0.30
    output / 1M
    $1.20
  • Venice

    venice/fp8 · fp8

    input / 1M
    $0.30
    output / 1M
    $1.20
  • Venice

    venice/ultrafast · fp8

    input / 1M
    $1.50
    output / 1M
    $6.00
  • Wafer

    wafer/mxfp4 · mxfp4

    input / 1M
    $0.19
    output / 1M
    $0.65
  • Wafer

    wafer/us · mxfp4

    input / 1M
    $0.051
    output / 1M
    $0.567
  • Google

    google-vertex

    input / 1M
    $0.30
    output / 1M
    $1.20

Use this model through Nautili

Route requests to DeepSeek V4.1 Flash through your Nautili gateway using a supported SDK or the Nautili SDK.

Connect your stack to Nautili

Route across supported providers through one control layer.