Skip to content

DeepSeek V4 Flash 0731

deepseek/deepseek-v4-flash-0731

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

Overview

DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows....

Providers

Access method depends on provider and account setup.

Pricing

  • Alibaba

    alibaba

    input / 1M
    $0.352
    output / 1M
    $1.056
  • Alibaba

    alibaba/us-east-1

    input / 1M
    $0.374
    output / 1M
    $1.122
  • AtlasCloud

    input / 1M
    $0.44
    output / 1M
    $1.32
  • Baidu

    input / 1M
    $0.44
    output / 1M
    $1.32
  • BaseTen

    input / 1M
    $0.13
    output / 1M
    $0.26
  • Cloudflare

    input / 1M
    $0.44
    output / 1M
    $1.32
  • Cohere

    input / 1M
    $0.14
    output / 1M
    $0.28
  • CoreWeave

    input / 1M
    $0.13
    output / 1M
    $0.28
  • DeepInfra

    input / 1M
    $0.06
    output / 1M
    $0.18
  • DigitalOcean

    input / 1M
    $0.119
    output / 1M
    $0.238
  • Fireworks

    input / 1M
    $0.22
    output / 1M
    $0.66
  • GMICloud

    input / 1M
    $0.286
    output / 1M
    $0.858
  • Inceptron

    input / 1M
    $0.1296
    output / 1M
    $0.60
  • Makora

    input / 1M
    $0.09
    output / 1M
    $0.195
  • Mancer 2

    input / 1M
    $0.20
    output / 1M
    $0.60
  • Morph

    morph

    input / 1M
    $0.142
    output / 1M
    $0.3996
  • Morph

    morph/bf16 · bf16

    input / 1M
    $0.142
    output / 1M
    $0.40
  • Nebius

    input / 1M
    $0.14
    output / 1M
    $0.28
  • NextBit

    input / 1M
    $0.352
    output / 1M
    $1.056
  • Novita

    input / 1M
    $0.4092
    output / 1M
    $1.2276
  • OpenInference

    open-inference/fp4 · fp4

    input / 1M
    $0.0144
    output / 1M
    $0.576
  • OpenInference

    open-inference/fp8 · fp8

    input / 1M
    $0.0128
    output / 1M
    $1.60
  • Parasail

    input / 1M
    $0.14
    output / 1M
    $0.28
  • Phala

    input / 1M
    $0.308
    output / 1M
    $0.924
  • Reka

    reka

    input / 1M
    $0.021
    output / 1M
    $0.528
  • Reka

    reka/fp4 · fp4

    input / 1M
    $0.088
    output / 1M
    $0.528
  • Relace

    input / 1M
    $0.016
    output / 1M
    $0.64
  • Sail Research

    sail-research/fp4 · fp4

    input / 1M
    $0.10
    output / 1M
    $0.30
  • Sail Research

    sail-research/us · fp4

    input / 1M
    $0.019
    output / 1M
    $0.42
  • SiliconFlow

    input / 1M
    $0.22
    output / 1M
    $0.66
  • StreamLake

    input / 1M
    $0.132
    output / 1M
    $0.396
  • Together

    input / 1M
    $0.14
    output / 1M
    $0.28
  • Venice

    input / 1M
    $0.13
    output / 1M
    $0.26
  • Wafer

    input / 1M
    $0.13
    output / 1M
    $0.23

Use this model through Nautili

Route requests to DeepSeek V4 Flash 0731 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.