Skip to content

DeepSeek V4 Flash 0731 (batch)

text → textHugging Face OpenRouter Cost calculator →

Best price · blended / 1M

$0.07

at Baidu

Input / output
$0.0466 / $0.14
Providers
24
Spread
7.4×
7 days
−20.0%
30 days
—
Context
1M
Prices per 1M tokens, blended: Baidu $0.07, DeepInfra (fp8/flex) $0.072, Relace $0.0875, DeepInfra $0.09, Wafer (us) $0.122, DeepInfra (fp8/priority) $0.135, DigitalOcean $0.149, BaseTen $0.163, BaseTen $0.163, Venice $0.163, Wafer (fast) $0.163, CoreWeave $0.165, Cohere $0.175, Nebius $0.175, Parasail $0.175, Reka $0.18, Sail Research $0.188, Sail Research (us) $0.188, StreamLake $0.198, Inceptron $0.221, Mancer 2 $0.238, Alibaba $0.264, Alibaba (us-east-1) $0.281, SiliconFlow $0.33, GMICloud $0.429, Novita $0.614, AtlasCloud $0.66, Chutes $0.66, Cloudflare $0.66, PrimeIntellect $0.66.

Providers

  • BaiduCheapest
    $0.07
    $0.0466 in · $0.14 out
    −21% vs ref
    1M ctx100% uptimefp8·−89% discount
  • fp8/flex
    $0.072
    $0.048 in · $0.144 out
    −19% vs ref
    1M ctx46.7% uptimefp8
  • $0.0875
    $0.07 in · $0.14 out
    −1.4% vs ref
    1M ctx99.9% uptimefp4
  • $0.09
    $0.06 in · $0.18 out
    +1.4% vs ref
    1M ctx98.8% uptimefp8
  • $0.122
    $0.062 in · $0.30 out
    +37% vs ref
    1M ctx100% uptimenvfp4
  • fp8/priority
    $0.135
    $0.09 in · $0.27 out
    +52% vs ref
    1M ctx99.4% uptimefp8
  • $0.149
    $0.119 in · $0.238 out
    +68% vs ref
    1M ctx100% uptime
  • $0.163
    $0.13 in · $0.26 out
    +83% vs ref
    1M ctx100% uptimefp32
  • $0.163
    $0.13 in · $0.26 out
    +83% vs ref
    1M ctx100% uptimefp8
  • $0.163
    $0.13 in · $0.26 out
    +83% vs ref
    1M ctx99.9% uptime
  • fast
    $0.163
    $0.14 in · $0.23 out
    +83% vs ref
    1M ctx100% uptimenvfp4
  • $0.165
    $0.13 in · $0.27 out
    +86% vs ref
    256K ctx100% uptimefp8
  • $0.175
    $0.14 in · $0.28 out
    +97% vs ref
    1M ctx100% uptimefp8
  • $0.175
    $0.14 in · $0.28 out
    +97% vs ref
    1M ctx100% uptimefp8
  • $0.175
    $0.14 in · $0.28 out
    +97% vs ref
    1M ctx99.9% uptimefp8
  • $0.18
    $0.10 in · $0.42 out
    +103% vs ref
    256K ctx100% uptime
  • $0.188
    $0.15 in · $0.30 out
    +111% vs ref
    1M ctx100% uptimefp4
  • $0.188
    $0.15 in · $0.30 out
    +111% vs ref
    1M ctx99.7% uptimefp4
  • $0.198
    $0.132 in · $0.396 out
    +123% vs ref
    1M ctx100% uptimefp8·−70% discount
  • $0.221
    $0.13 in · $0.495 out
    +149% vs ref
    1M ctx99.9% uptimefp4
  • $0.238
    $0.15 in · $0.50 out
    +168% vs ref
    1M ctx98.6% uptimefp8
  • $0.264
    $0.176 in · $0.528 out
    +197% vs ref
    1M ctx99.8% uptime
  • us-east-1
    $0.281
    $0.187 in · $0.561 out
    +216% vs ref
    1M ctx100% uptime
  • $0.33
    $0.22 in · $0.66 out
    +272% vs ref
    1M ctx99.3% uptimefp8
  • $0.429
    $0.286 in · $0.858 out
    +383% vs ref
    1M ctx100% uptimefp8·−35% discount
  • $0.614
    $0.409 in · $1.23 out
    +592% vs ref
    1M ctx100% uptimefp8·−7% discount
  • $0.66
    $0.44 in · $1.32 out
    +644% vs ref
    1M ctx100% uptimefp4
  • $0.66
    $0.44 in · $1.32 out
    +644% vs ref
    1M ctx100% uptimefp8
  • $0.66
    $0.44 in · $1.32 out
    +644% vs ref
    1M ctx100% uptime
  • $0.66
    $0.44 in · $1.32 out
    +644% vs ref
    1M ctx100% uptimefp4
Blended = (3 × input + output) / 4. Ref $0.0888 = median of cheapest 20%.

Price history

USD per 1M tokens · each step is a price change

Prices in USD / 1M tokens.

Long-context pricing

Offered byPrompt sizeInputOutput
Alibababase$0.176$0.528
> 0$0.352$1.056
> 0$0.176$0.528
Alibaba (us-east-1)base$0.187$0.561
> 0$0.374$1.122
> 0$0.187$0.561

Other pricing

Offered byCache read
Baidu$0.00148
DeepInfra (fp8/flex)$0.012
Relace, Cloudflare +1$0.014
DeepInfra$0.015
Wafer (us)$0.035
DeepInfra (fp8/priority)$0.0225
DigitalOcean$0.0238
BaseTen, BaseTen +3$0.028
Wafer (fast)$0.135
CoreWeave, Parasail$0.05
Cohere$0.07
Nebius$0.01
Reka$0.06
Sail Research, Sail Research (us)$0.042
StreamLake$0.0042
Inceptron$0.027
Mancer 2$0.075
Alibaba$0.0176
Alibaba (us-east-1)$0.0187
GMICloud$0.0091
Novita$0.026
Chutes$0.044

Changes

  • Cache read price,
    $0.0374→$0.0187−50.0%
  • Output price,
    $1.122→$0.561−50.0%
  • Input price,
    $0.374→$0.187−50.0%
  • Cache read price,
    $0.0352→$0.0176−50.0%
  • Output price,
    $1.056→$0.528−50.0%
  • Input price,
    $0.352→$0.176−50.0%
  • Discount price,
    —→89%—
  • Cache read price,
    $0.014→$0.00148−89.4%
  • Output price,
    $1.32→$0.14−89.4%
  • Input price,
    $0.44→$0.0466−89.4%
  • Input price,
    $0.055→$0.062+12.7%
  • Cache read price,
    $0.0187→$0.0374+100%
  • Output price,
    $0.561→$1.122+100%
  • Input price,
    $0.187→$0.374+100%
  • Cache read price,
    $0.0176→$0.0352+100%

About

The cheapest provider for DeepSeek V4 Flash 0731 (batch) is Baidu at $0.07 per 1M tokens (blended), charging $0.0466 per 1M input and $0.14 per 1M output tokens. 24 providers serve it; the most expensive charges 7.4× the typical price. Context window: 1M tokens. Prices tracked since Nov 7, 2026.

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....
Released
Jul 31, 2026
First tracked
Nov 7, 2026
Canonical slug
deepseek/deepseek-v4-flash-20260731
Weights
deepseek-ai/DeepSeek-V4-Flash-0731