Skip to content

Compare model prices

Input, output, cached and batch prices, every route and a month of tokens, side by side.

DeepSeek V3.2DeepSeek V4.1 Flash

A month of tokens

  • DeepSeek V3.2$27.07
  • DeepSeek V4.1 Flash$27.00

100M input and 20M output tokens, list price. Use your own numbers.

Side by side

DeepSeek V3.2DeepSeek V4.1 Flash
Input per 1M$0.209$0.15
Output per 1M$0.31$0.6
Cached input$0.022$0.003
Batch in / outn/a$0.112 / $0.336
Blended (3:1)$0.234$0.262
Context164K1M
Routes1525
ReleasedDec 1, 2025Sep 10, 2026
Leadersboard scoren/a3 (#33)

On the price map

$0.03$0.1$0.3$1$3$10$30$100$0.03$0.1$0.3$1$3$10$30$100$300Input $ per 1M tokensOutput $ per 1M tokensxxZZDMOOOOOOOOOOOOOOOOZDGMMDMGDQMGGMMZGGDDGGGGGGGGMOOOZQAQQAQQQAAAQAQxQQAQQAQAAAAQQQQxxxOOOOOOOOOOOOOOOOOAAOMMQQQMGGMQQQQQQQMMMMQQQQQQQMQQQQxMMCZZZZZZZAAAOAOOOOOOOACDDDDDMDDMGGZMOQQQMOOZZQMMQQQQQGGGMMMMMMOOOMQCAOCCOOOOOOMQGGGZMDMMGMMQQQ
Size: context window, 8K to 2MEqual blended cost (3:1) 243 models, lower left is cheaper

Every route

DeepSeek V3.2

  • GMICloudfp8, fp8$0.209 / $0.31cached $0.022
  • StreamLakefp8, fp8+3%$0.215 / $0.322cached $0.021
  • AtlasCloudfp8, fp8+24%$0.26 / $0.38cached $0.13
  • DeepInfrafp4, fp4+24%$0.26 / $0.38cached $0.13
  • Venicedefault region+28%$0.268 / $0.39cached $0.13
  • SiliconFlowfp8, fp8+28%$0.259 / $0.42cached $0.135
  • Novita AIfp8, fp8+29%$0.269 / $0.4cached $0.135
  • Baidufp8, fp8+35%$0.28 / $0.42cached $0.028
  • DigitalOceandefault region+99%$0.3 / $0.96cached $0.09
  • Qwen (Alibaba)fp8, fp8+137%$0.371 / $1.11cached $0.074
  • Friendlidefault region+221%$0.5 / $1.5cached $0.25
  • Google Vertex AIdefault region+259%$0.56 / $1.68$0.84 blended
  • Phaladefault region+327%$1 / $1cached $0.5
  • Maradefault region+1342%$3 / $4.5$3.38 blended
  • SambaNovadefault region+1342%$3 / $4.5$3.38 blended

DeepSeek V4.1 Flash

  • Relacedefault region-62%$0.05 / $0.25cached $0.005
  • Morphdefault region-52%$0.072 / $0.288cached $0.0014
  • OpenInferencefp4, fp4-24%$0.1 / $0.5cached $0.01
  • DeepInfrafp8, fp8-20%$0.14 / $0.42cached $0.0042
  • DeepSeek (list)the lab's own API$0.15 / $0.6cached $0.003
  • Sail Researchfp4, fp4+9%$0.13 / $0.75cached $0.01
  • Waferdefault region+14%$0.2 / $0.6cached $0.006
  • CoreWeavefp8, fp8+19%$0.2 / $0.65cached $0.03
  • DekaLLMdefault region+24%$0.1 / $1cached $0.01
  • Fireworks AIdefault region+26%$0.22 / $0.66cached $0.007
  • StreamLakefp8, fp8+32%$0.199 / $0.794cached $0.004
  • GMICloudfp8, fp8+50%$0.225 / $0.9cached $0.0045
  • Kreafp8, fp8+50%$0.225 / $0.9cached $0.006
  • Phaladefault region+84%$0.276 / $1.1cached $0.0055
  • Novita AIfp8, fp8+90%$0.285 / $1.14cached $0.0057
  • AtlasCloudfp8, fp8+100%$0.3 / $1.2cached $0.03
  • Basetenfp8, fp8+100%$0.3 / $1.2cached $0.03
  • DigitalOceandefault region+100%$0.3 / $1.2cached $0.006
  • Makorafp8, fp8+100%$0.3 / $1.2cached $0.006
  • Modaldefault region+100%$0.3 / $1.2cached $0.03
  • Parasailfp8, fp8+100%$0.3 / $1.2cached $0.006
  • Qwen (Alibaba)default region+100%$0.3 / $1.2cached $0.03
  • SiliconFlowfp8, fp8+100%$0.3 / $1.2cached $0.006
  • Together AIdefault region+100%$0.3 / $1.2cached $0.006
  • Venicefp8, fp8+150%$0.375 / $1.5cached $0.0075

List prices from providers' pages on the date shown; your contract may differ. Logos via logo.dev; trademarks belong to their owners.

Weekly: LLM API list prices that changed, Thursday mornings.