{
 "metadata": {
  "name": "DeepSeek API pricing as stated by DeepSeek",
  "source": "https://api-docs.deepseek.com/quick_start/pricing",
  "archivedFile": "deepseek-models-and-pricing.rendered.txt",
  "archivedSha256": "dfb3d0a8006d276f325246cbff87cc402f3d3e7df135be6d41c18506f4e71b40",
  "retrievedUtc": "2026-10-26T09:47:53.636Z",
  "unit": "USD per 1,000,000 tokens",
  "limits": [
   "Read from one page on one date; DeepSeek says it may adjust prices.",
   "OpenRouter host prices are a separate source (dispersion.json); hosts differ in quantisation, context, speed, uptime and data policy."
  ]
 },
 "models": {
  "flash": {
   "apiName": "deepseek-flash",
   "version": "DeepSeek-V4.1-Flash-1017"
  },
  "pro": {
   "apiName": "deepseek-v4-pro",
   "version": "DeepSeek-V4-Pro-0813"
  },
  "contextLength": "1M",
  "maxOutput": "384K",
  "concurrency": {
   "flash": 2500,
   "pro": 500
  }
 },
 "price": {
  "cacheHit": {
   "offPeak": {
    "flash": 0.003,
    "pro": 0.022
   },
   "peak": {
    "flash": 0.006,
    "pro": 0.044
   }
  },
  "cacheMiss": {
   "offPeak": {
    "flash": 0.15,
    "pro": 0.66
   },
   "peak": {
    "flash": 0.3,
    "pro": 1.32
   }
  },
  "output": {
   "offPeak": {
    "flash": 0.6,
    "pro": 1.98
   },
   "peak": {
    "flash": 1.2,
    "pro": 3.96
   }
  }
 },
 "peakWindowsUtc": [
  [
   1,
   4
  ],
  [
   6,
   10
  ]
 ],
 "peakHoursPerWeekday": 7,
 "peakShareOfWeekHours": 20.8,
 "blended": {
  "flash": {
   "offPeak": 0.2625,
   "peak": 0.525
  },
  "pro": {
   "offPeak": 0.99,
   "peak": 1.98
  }
 },
 "searchFetch": {
  "searchPer1000": 0.75,
  "fetchPer1000": 0.75,
  "billingStartsBeijing": "2026-11-01T00:00+08:00",
  "freeUntil": "billing start; free today when called through DeepSeek Harness, Claude Code or Codex"
 },
 "hostsFlash": {
  "model": "deepseek/deepseek-v4.1-flash-1017",
  "hosts": 30,
  "spreadRatio": 5.1,
  "endpoints": [
   {
    "host": "Sail Research",
    "input": 0.101,
    "output": 0.15,
    "blended": 0.11325,
    "quantization": "fp4"
   },
   {
    "host": "DeepInfra",
    "input": 0.0784,
    "output": 0.2352,
    "blended": 0.1176,
    "quantization": "fp8"
   },
   {
    "host": "Reka",
    "input": 0.04,
    "output": 0.5,
    "blended": 0.155,
    "quantization": "unknown"
   },
   {
    "host": "Relace",
    "input": 0.07,
    "output": 0.5,
    "blended": 0.1775,
    "quantization": "unknown"
   },
   {
    "host": "InferenceNet",
    "input": 0.08,
    "output": 0.5,
    "blended": 0.185,
    "quantization": "fp8"
   },
   {
    "host": "Wafer",
    "input": 0.13,
    "output": 0.5,
    "blended": 0.2225,
    "quantization": "mxfp4"
   },
   {
    "host": "Cloudflare",
    "input": 0.1,
    "output": 0.6,
    "blended": 0.225,
    "quantization": "unknown"
   },
   {
    "host": "Morph",
    "input": 0.1,
    "output": 0.6,
    "blended": 0.225,
    "quantization": "fp8"
   },
   {
    "host": "Decart",
    "input": 0.1365,
    "output": 0.546,
    "blended": 0.238875,
    "quantization": "fp4"
   },
   {
    "host": "OpenInference",
    "input": 0.02,
    "output": 1,
    "blended": 0.265,
    "quantization": "fp4"
   },
   {
    "host": "GMICloud",
    "input": 0.18,
    "output": 0.72,
    "blended": 0.315,
    "quantization": "fp8"
   },
   {
    "host": "Ionstream",
    "input": 0.22,
    "output": 0.6,
    "blended": 0.315,
    "quantization": "unknown"
   },
   {
    "host": "AtlasCloud",
    "input": 0.195,
    "output": 0.78,
    "blended": 0.34125,
    "quantization": "fp8"
   },
   {
    "host": "Fireworks",
    "input": 0.21,
    "output": 0.84,
    "blended": 0.3675,
    "quantization": "unknown"
   },
   {
    "host": "StreamLake",
    "input": 0.21,
    "output": 0.84,
    "blended": 0.3675,
    "quantization": "fp8"
   },
   {
    "host": "DigitalOcean",
    "input": 0.225,
    "output": 0.9,
    "blended": 0.39375,
    "quantization": "unknown"
   },
   {
    "host": "Makora",
    "input": 0.15,
    "output": 1.2,
    "blended": 0.4125,
    "quantization": "fp8"
   },
   {
    "host": "Novita",
    "input": 0.24,
    "output": 0.96,
    "blended": 0.42,
    "quantization": "fp8"
   },
   {
    "host": "Modal",
    "input": 0.25,
    "output": 1,
    "blended": 0.4375,
    "quantization": "unknown"
   },
   {
    "host": "Baidu",
    "input": 0.28,
    "output": 1.13,
    "blended": 0.4925,
    "quantization": "fp8"
   },
   {
    "host": "Crusoe",
    "input": 0.29,
    "output": 1.2,
    "blended": 0.5175,
    "quantization": "fp8"
   },
   {
    "host": "Nebius",
    "input": 0.3,
    "output": 1.2,
    "blended": 0.525,
    "quantization": "fp8"
   },
   {
    "host": "Phala",
    "input": 0.3,
    "output": 1.2,
    "blended": 0.525,
    "quantization": "unknown"
   },
   {
    "host": "SiliconFlow",
    "input": 0.3,
    "output": 1.2,
    "blended": 0.525,
    "quantization": "fp8"
   },
   {
    "host": "Together",
    "input": 0.3,
    "output": 1.2,
    "blended": 0.525,
    "quantization": "unknown"
   },
   {
    "host": "PrimeIntellect",
    "input": 0.3,
    "output": 1.2,
    "blended": 0.525,
    "quantization": "unknown"
   },
   {
    "host": "BaseTen",
    "input": 0.3,
    "output": 1.2,
    "blended": 0.525,
    "quantization": "unknown"
   },
   {
    "host": "Parasail",
    "input": 0.3,
    "output": 1.2,
    "blended": 0.525,
    "quantization": "fp8"
   },
   {
    "host": "DeepSeek",
    "input": 0.3,
    "output": 1.2,
    "blended": 0.525,
    "quantization": "unknown"
   },
   {
    "host": "Alibaba",
    "input": 0.33,
    "output": 1.32,
    "blended": 0.5775,
    "quantization": "unknown"
   }
  ]
 },
 "hostsPro": {
  "model": "deepseek/deepseek-v4-pro-0813",
  "hosts": 22,
  "spreadRatio": 4.6,
  "endpoints": [
   {
    "host": "Sail Research",
    "input": 0.45,
    "output": 0.8,
    "blended": 0.5375,
    "quantization": "fp4"
   },
   {
    "host": "StreamLake",
    "input": 0.66,
    "output": 1.98,
    "blended": 0.99,
    "quantization": "unknown"
   },
   {
    "host": "Ionstream",
    "input": 0.37,
    "output": 2.93,
    "blended": 1.01,
    "quantization": "unknown"
   },
   {
    "host": "Wafer",
    "input": 0.285,
    "output": 4.3,
    "blended": 1.28875,
    "quantization": "mxfp4"
   },
   {
    "host": "Novita",
    "input": 0.99,
    "output": 2.97,
    "blended": 1.485,
    "quantization": "fp8"
   },
   {
    "host": "GMICloud",
    "input": 1.056,
    "output": 3.168,
    "blended": 1.584,
    "quantization": "fp8"
   },
   {
    "host": "DeepInfra",
    "input": 1.3,
    "output": 2.6,
    "blended": 1.625,
    "quantization": "fp8"
   },
   {
    "host": "Alibaba",
    "input": 1.122,
    "output": 3.366,
    "blended": 1.683,
    "quantization": "unknown"
   },
   {
    "host": "DekaLLM",
    "input": 1.08,
    "output": 3.8,
    "blended": 1.76,
    "quantization": "unknown"
   },
   {
    "host": "CoreWeave",
    "input": 1.31,
    "output": 3.96,
    "blended": 1.9725,
    "quantization": "fp8"
   },
   {
    "host": "BaseTen",
    "input": 1.32,
    "output": 3.96,
    "blended": 1.98,
    "quantization": "fp4"
   },
   {
    "host": "AtlasCloud",
    "input": 1.32,
    "output": 3.96,
    "blended": 1.98,
    "quantization": "fp8"
   },
   {
    "host": "Baidu",
    "input": 1.32,
    "output": 3.96,
    "blended": 1.98,
    "quantization": "fp8"
   },
   {
    "host": "Parasail",
    "input": 1.32,
    "output": 3.96,
    "blended": 1.98,
    "quantization": "fp8"
   },
   {
    "host": "Together",
    "input": 1.32,
    "output": 3.96,
    "blended": 1.98,
    "quantization": "unknown"
   },
   {
    "host": "DigitalOcean",
    "input": 1.32,
    "output": 3.96,
    "blended": 1.98,
    "quantization": "unknown"
   },
   {
    "host": "SiliconFlow",
    "input": 1.32,
    "output": 3.96,
    "blended": 1.98,
    "quantization": "fp8"
   },
   {
    "host": "Cloudflare",
    "input": 1.32,
    "output": 3.96,
    "blended": 1.98,
    "quantization": "unknown"
   },
   {
    "host": "DeepSeek",
    "input": 1.32,
    "output": 3.96,
    "blended": 1.98,
    "quantization": "unknown"
   },
   {
    "host": "Phala",
    "input": 1.45,
    "output": 4.36,
    "blended": 2.1775,
    "quantization": "unknown"
   },
   {
    "host": "Relace",
    "input": 0.2565,
    "output": 8,
    "blended": 2.192375,
    "quantization": "fp4"
   },
   {
    "host": "Venice",
    "input": 1.65,
    "output": 4.95,
    "blended": 2.475,
    "quantization": "unknown"
   }
  ]
 }
}
