{
 "metadata": {
  "name": "Same Weights, Different Price: what hosts charge for identical open-weights LLMs",
  "generated": "2026-10-26T09:18:17.868Z",
  "source": "OpenRouter public API (models, and per-model endpoints), read on the generation date",
  "unit": "USD per 1,000,000 tokens, list price per host. Blended = (3 x input + output) / 4, a stated convention.",
  "scope": "open-weights models (have a Hugging Face id) with at least 3 priced hosts on OpenRouter. Only hosts listed on OpenRouter.",
  "limits": [
   "Hosts differ in quantisation, context length, speed, uptime and data policy. A price gap is not evidence of overcharging.",
   "sameQuantisationEverywhere = true narrows one confounder only; speed, context and policy still differ.",
   "One row per host: where OpenRouter lists several endpoints for a host, the cheapest is kept (the best price a buyer could get there).",
   "OpenRouter lists the price it routes at; a host may price differently direct. Read on one date: prices move.",
   "Snapshot, not yet a series. Monthly re-reads make it one."
  ]
 },
 "summary": {
  "modelsWithAtLeast3Hosts": 65,
  "endpointsProbedOk": 156,
  "endpointsFailed": 0,
  "medianSpreadAll": 2.05,
  "medianSpreadSameQuantisation": 1.9,
  "modelsWithSameQuantisation": 3,
  "modelsSpread5xOrMore": 14,
  "fp8Only": {
   "models": 36,
   "medianSpread": 1.82,
   "p90Spread": 4.53,
   "maxSpread": 7.43
  }
 },
 "models": [
  {
   "model": "deepseek/deepseek-v3.2",
   "huggingFaceId": "deepseek-ai/DeepSeek-V3.2",
   "hosts": 11,
   "cheapest": {
    "host": "GMICloud",
    "input": 0.2088,
    "output": 0.3096,
    "quantization": "fp8",
    "contextLength": 163840,
    "uptime30m": 100,
    "blended": 0.234,
    "variants": 1
   },
   "dearest": {
    "host": "SambaNova",
    "input": 3,
    "output": 4.5,
    "quantization": "unknown",
    "contextLength": 32768,
    "uptime30m": 100,
    "blended": 3.375,
    "variants": 1
   },
   "spreadRatio": 14.42,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "GMICloud",
     "input": 0.2088,
     "output": 0.3096,
     "quantization": "fp8",
     "contextLength": 163840,
     "uptime30m": 100,
     "blended": 0.234,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.26,
     "output": 0.38,
     "quantization": "fp4",
     "contextLength": 163840,
     "uptime30m": 93.33333333333333,
     "blended": 0.29,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.26,
     "output": 0.38,
     "quantization": "fp8",
     "contextLength": 163840,
     "uptime30m": 100,
     "blended": 0.29,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.26829,
     "output": 0.39024,
     "quantization": "unknown",
     "contextLength": 160000,
     "uptime30m": 100,
     "blended": 0.298778,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 0.28,
     "output": 0.42,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": null,
     "blended": 0.315,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.259,
     "output": 0.8,
     "quantization": "fp8",
     "contextLength": 163840,
     "uptime30m": 100,
     "blended": 0.39425,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.3,
     "output": 0.96,
     "quantization": "unknown",
     "contextLength": 163840,
     "uptime30m": 100,
     "blended": 0.465,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.56,
     "output": 1.68,
     "quantization": "unknown",
     "contextLength": 163840,
     "uptime30m": 0,
     "blended": 0.84,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 1,
     "output": 1,
     "quantization": "unknown",
     "contextLength": 163840,
     "uptime30m": 100,
     "blended": 1,
     "variants": 1
    },
    {
     "host": "Mara",
     "input": 3,
     "output": 4.5,
     "quantization": "unknown",
     "contextLength": 32768,
     "uptime30m": 0,
     "blended": 3.375,
     "variants": 1
    },
    {
     "host": "SambaNova",
     "input": 3,
     "output": 4.5,
     "quantization": "unknown",
     "contextLength": 32768,
     "uptime30m": 100,
     "blended": 3.375,
     "variants": 1
    }
   ]
  },
  {
   "model": "z-ai/glm-5.3-flash",
   "huggingFaceId": "zai-org/GLM-5.3-Flash",
   "hosts": 34,
   "cheapest": {
    "host": "Relace",
    "input": 0.0413,
    "output": 0.0758,
    "quantization": "unknown",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.049925,
    "variants": 1
   },
   "dearest": {
    "host": "Reka",
    "input": 0.15,
    "output": 2.4,
    "quantization": "unknown",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.7125,
    "variants": 1
   },
   "spreadRatio": 14.27,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Relace",
     "input": 0.0413,
     "output": 0.0758,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.049925,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.0435,
     "output": 0.145,
     "quantization": "fp8",
     "contextLength": 1024000,
     "uptime30m": 80,
     "blended": 0.068875,
     "variants": 1
    },
    {
     "host": "OpenInference",
     "input": 0.110038,
     "output": 0.01,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.085029,
     "variants": 1
    },
    {
     "host": "Wafer",
     "input": 0.054,
     "output": 0.2,
     "quantization": "nvfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.0905,
     "variants": 1
    },
    {
     "host": "Darkbloom",
     "input": 0.065,
     "output": 0.18,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 0.09375,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.075,
     "output": 0.25,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.11875,
     "variants": 1
    },
    {
     "host": "Sail Research",
     "input": 0.07835,
     "output": 0.25,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 0.121263,
     "variants": 2
    },
    {
     "host": "Novita",
     "input": 0.084,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.133,
     "variants": 1
    },
    {
     "host": "Near AI",
     "input": 0.08,
     "output": 0.5,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.185,
     "variants": 1
    },
    {
     "host": "InferenceNet",
     "input": 0.12,
     "output": 0.388,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.187,
     "variants": 2
    },
    {
     "host": "Phala",
     "input": 0.1275,
     "output": 0.425,
     "quantization": "nvfp4",
     "contextLength": 1048576,
     "uptime30m": 96.66666666666667,
     "blended": 0.201875,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.13125,
     "output": 0.4375,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.207813,
     "variants": 1
    },
    {
     "host": "NextBit",
     "input": 0.125,
     "output": 0.5,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.21875,
     "variants": 1
    },
    {
     "host": "Makora",
     "input": 0.14,
     "output": 0.47,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2225,
     "variants": 1
    },
    {
     "host": "Mistral",
     "input": 0.15,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Cohere",
     "input": 0.15,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "PrimeIntellect",
     "input": 0.15,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 0.15,
     "output": 0.5,
     "quantization": "fp4",
     "contextLength": 1024000,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Modal",
     "input": 0.15,
     "output": 0.5,
     "quantization": "nvfp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 0.15,
     "output": 0.5,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 0.2375,
     "variants": 2
    },
    {
     "host": "Crusoe",
     "input": 0.15,
     "output": 0.5,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.15,
     "output": 0.5,
     "quantization": "nvfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.15,
     "output": 0.5,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Fireworks",
     "input": 0.15,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 0.2375,
     "variants": 2
    },
    {
     "host": "Friendli",
     "input": 0.15,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.15,
     "output": 0.5,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 0.15,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048575,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.15,
     "output": 0.5,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.15,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.15,
     "output": 0.5,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Z.AI",
     "input": 0.15,
     "output": 0.5,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Inceptron",
     "input": 0.15,
     "output": 0.55,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.25,
     "variants": 1
    },
    {
     "host": "Morph",
     "input": 0.199,
     "output": 0.65,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.31175,
     "variants": 1
    },
    {
     "host": "Reka",
     "input": 0.15,
     "output": 2.4,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.7125,
     "variants": 1
    }
   ]
  },
  {
   "model": "deepseek/deepseek-v4-flash-0731",
   "huggingFaceId": "deepseek-ai/DeepSeek-V4-Flash-0731",
   "hosts": 22,
   "cheapest": {
    "host": "Relace",
    "input": 0.0475,
    "output": 0.1568,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.074825,
    "variants": 1
   },
   "dearest": {
    "host": "Cloudflare",
    "input": 0.44,
    "output": 1.32,
    "quantization": "unknown",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.66,
    "variants": 1
   },
   "spreadRatio": 8.82,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Relace",
     "input": 0.0475,
     "output": 0.1568,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.074825,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.06,
     "output": 0.18,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.09,
     "variants": 1
    },
    {
     "host": "Reka",
     "input": 0.05,
     "output": 0.28,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 86.36363636363636,
     "blended": 0.1075,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.119,
     "output": 0.238,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.14875,
     "variants": 1
    },
    {
     "host": "Sail Research",
     "input": 0.15,
     "output": 0.15,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.15,
     "variants": 2
    },
    {
     "host": "Wafer",
     "input": 0.13,
     "output": 0.23,
     "quantization": "nvfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.155,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 0.13,
     "output": 0.26,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.1625,
     "variants": 2
    },
    {
     "host": "Venice",
     "input": 0.13,
     "output": 0.26,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.1625,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.13,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1675,
     "variants": 1
    },
    {
     "host": "Cohere",
     "input": 0.14,
     "output": 0.28,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.14,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "Inceptron",
     "input": 0.11,
     "output": 0.4125,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.185625,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.132,
     "output": 0.396,
     "quantization": "fp8",
     "contextLength": 1024000,
     "uptime30m": 100,
     "blended": 0.198,
     "variants": 1
    },
    {
     "host": "Mancer 2",
     "input": 0.2,
     "output": 0.6,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.3,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.22,
     "output": 0.66,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.33,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.286,
     "output": 0.858,
     "quantization": "fp8",
     "contextLength": 1048575,
     "uptime30m": 100,
     "blended": 0.429,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.352,
     "output": 1.056,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 92.5925925925926,
     "blended": 0.528,
     "variants": 2
    },
    {
     "host": "Novita",
     "input": 0.4092,
     "output": 1.2276,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.6138,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.44,
     "output": 1.32,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.66,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 0.44,
     "output": 1.32,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.66,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.44,
     "output": 1.32,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.66,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 0.44,
     "output": 1.32,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.66,
     "variants": 1
    }
   ]
  },
  {
   "model": "meta-llama/llama-3.1-8b-instruct",
   "huggingFaceId": "meta-llama/Meta-Llama-3.1-8B-Instruct",
   "hosts": 5,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.02,
    "output": 0.04,
    "quantization": "fp8",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.025,
    "variants": 1
   },
   "dearest": {
    "host": "CoreWeave",
    "input": 0.22,
    "output": 0.22,
    "quantization": "bf16",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.22,
    "variants": 1
   },
   "spreadRatio": 8.8,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.02,
     "output": 0.04,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.025,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.02,
     "output": 0.05,
     "quantization": "fp8",
     "contextLength": 16384,
     "uptime30m": 100,
     "blended": 0.0275,
     "variants": 1
    },
    {
     "host": "Groq",
     "input": 0.05,
     "output": 0.08,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.0575,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 0.152,
     "output": 0.287,
     "quantization": "fp8",
     "contextLength": 32000,
     "uptime30m": 100,
     "blended": 0.18575,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.22,
     "output": 0.22,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.22,
     "variants": 1
    }
   ]
  },
  {
   "model": "mistralai/mistral-nemo",
   "huggingFaceId": "mistralai/Mistral-Nemo-Instruct-2407",
   "hosts": 5,
   "cheapest": {
    "host": "Io Net",
    "input": 0.01656,
    "output": 0.0216,
    "quantization": "fp16",
    "contextLength": 128000,
    "uptime30m": 100,
    "blended": 0.01782,
    "variants": 1
   },
   "dearest": {
    "host": "Mistral",
    "input": 0.15,
    "output": 0.15,
    "quantization": "unknown",
    "contextLength": 131072,
    "uptime30m": null,
    "blended": 0.15,
    "variants": 1
   },
   "spreadRatio": 8.42,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Io Net",
     "input": 0.01656,
     "output": 0.0216,
     "quantization": "fp16",
     "contextLength": 128000,
     "uptime30m": 100,
     "blended": 0.01782,
     "variants": 1
    },
    {
     "host": "DekaLLM",
     "input": 0.018,
     "output": 0.03,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.021,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.029,
     "output": 0.03,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.02925,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.03,
     "output": 0.03,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.03,
     "variants": 1
    },
    {
     "host": "Mistral",
     "input": 0.15,
     "output": 0.15,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": null,
     "blended": 0.15,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3.8-27b",
   "huggingFaceId": "Qwen/Qwen3.8-27B",
   "hosts": 21,
   "cheapest": {
    "host": "Tenstorrent",
    "input": 0.02,
    "output": 0.5,
    "quantization": "fp4",
    "contextLength": 262144,
    "uptime30m": 52.17391304347826,
    "blended": 0.14,
    "variants": 1
   },
   "dearest": {
    "host": "Venice",
    "input": 0.45,
    "output": 3.2,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 1.1375,
    "variants": 1
   },
   "spreadRatio": 8.12,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Tenstorrent",
     "input": 0.02,
     "output": 0.5,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 52.17391304347826,
     "blended": 0.14,
     "variants": 1
    },
    {
     "host": "Chutes",
     "input": 0.039,
     "output": 1.25,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.34175,
     "variants": 1
    },
    {
     "host": "Near AI",
     "input": 0.04,
     "output": 1.35,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.3675,
     "variants": 1
    },
    {
     "host": "Ionstream",
     "input": 0.04,
     "output": 1.95,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.5175,
     "variants": 1
    },
    {
     "host": "Darkbloom",
     "input": 0.035,
     "output": 2.2,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.57625,
     "variants": 1
    },
    {
     "host": "Reka",
     "input": 0.04,
     "output": 2.2,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.58,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.15,
     "output": 1.875,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.58125,
     "variants": 1
    },
    {
     "host": "Modal",
     "input": 0.04,
     "output": 2.3,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.605,
     "variants": 1
    },
    {
     "host": "Wafer",
     "input": 0.06,
     "output": 2.3,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.62,
     "variants": 1
    },
    {
     "host": "DekaLLM",
     "input": 0.04,
     "output": 2.5,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.655,
     "variants": 1
    },
    {
     "host": "AkashML",
     "input": 0.225,
     "output": 1.98,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.66375,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.24,
     "output": 2.2,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.73,
     "variants": 1
    },
    {
     "host": "Mancer 2",
     "input": 0.18,
     "output": 2.5,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.76,
     "variants": 1
    },
    {
     "host": "ModelRun",
     "input": 0.2,
     "output": 2.5,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.775,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.2,
     "output": 2.5,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.775,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.425,
     "output": 2.55,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.95625,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.4,
     "output": 3,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.05,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.42,
     "output": 3,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 1.065,
     "variants": 1
    },
    {
     "host": "Cerebras",
     "input": 0.99,
     "output": 1.49,
     "quantization": "fp16",
     "contextLength": 65536,
     "uptime30m": 100,
     "blended": 1.115,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 0.45,
     "output": 3.2,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 1.1375,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.45,
     "output": 3.2,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.1375,
     "variants": 1
    }
   ]
  },
  {
   "model": "z-ai/glm-5.3",
   "huggingFaceId": "zai-org/GLM-5.3",
   "hosts": 37,
   "cheapest": {
    "host": "Wafer",
    "input": 0.0699,
    "output": 1.4,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.402425,
    "variants": 2
   },
   "dearest": {
    "host": "ModelRun",
    "input": 2.1,
    "output": 6.6,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": null,
    "blended": 3.225,
    "variants": 1
   },
   "spreadRatio": 8.01,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Wafer",
     "input": 0.0699,
     "output": 1.4,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.402425,
     "variants": 2
    },
    {
     "host": "Sail Research",
     "input": 0.5,
     "output": 2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.875,
     "variants": 2
    },
    {
     "host": "DekaLLM",
     "input": 0.25,
     "output": 2.95,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.925,
     "variants": 1
    },
    {
     "host": "InferenceNet",
     "input": 0.08,
     "output": 3.6,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.96,
     "variants": 2
    },
    {
     "host": "Novita",
     "input": 0.63,
     "output": 1.98,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.9675,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.5625,
     "output": 2.5,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.046875,
     "variants": 1
    },
    {
     "host": "Tenstorrent",
     "input": 0.6,
     "output": 2.5,
     "quantization": "fp8",
     "contextLength": 1044480,
     "uptime30m": 100,
     "blended": 1.075,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.7,
     "output": 2.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.075,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.7,
     "output": 2.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.075,
     "variants": 1
    },
    {
     "host": "Relace",
     "input": 0.0703,
     "output": 4.3843,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.1488,
     "variants": 1
    },
    {
     "host": "Inceptron",
     "input": 0.6,
     "output": 2.8,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.15,
     "variants": 1
    },
    {
     "host": "Reka",
     "input": 0.14,
     "output": 4.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.155,
     "variants": 1
    },
    {
     "host": "Makora",
     "input": 0.14,
     "output": 4.4,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.205,
     "variants": 1
    },
    {
     "host": "AkashML",
     "input": 0.19,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.2425,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.84,
     "output": 2.64,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.29,
     "variants": 1
    },
    {
     "host": "Decart",
     "input": 0.84,
     "output": 2.71,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.3075,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.91,
     "output": 2.86,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.3975,
     "variants": 1
    },
    {
     "host": "Fireworks",
     "input": 0.98,
     "output": 3.08,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.505,
     "variants": 3
    },
    {
     "host": "PrimeIntellect",
     "input": 0.98,
     "output": 3.08,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.505,
     "variants": 1
    },
    {
     "host": "Morph",
     "input": 0.669,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 83.33333333333334,
     "blended": 1.60175,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 1.19,
     "output": 3.74,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": null,
     "blended": 1.8275,
     "variants": 2
    },
    {
     "host": "CoreWeave",
     "input": 1.2,
     "output": 4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.9,
     "variants": 1
    },
    {
     "host": "Friendli",
     "input": 1.26,
     "output": 3.96,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.935,
     "variants": 1
    },
    {
     "host": "Mistral",
     "input": 1.4,
     "output": 4.4,
     "quantization": "nvfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 3
    },
    {
     "host": "Baidu",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 4
    },
    {
     "host": "Nebius",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp4",
     "contextLength": 1024000,
     "uptime30m": null,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Crusoe",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 1.4,
     "output": 4.4,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": null,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 1.4,
     "output": 4.4,
     "quantization": "unknown",
     "contextLength": 1048575,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Modal",
     "input": 1.4,
     "output": 4.4,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 1.4,
     "output": 4.4,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Z.AI",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Amazon Bedrock",
     "input": 1.68,
     "output": 5.28,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.58,
     "variants": 1
    },
    {
     "host": "ModelRun",
     "input": 2.1,
     "output": 6.6,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 3.225,
     "variants": 1
    }
   ]
  },
  {
   "model": "openai/gpt-oss-120b",
   "huggingFaceId": "openai/gpt-oss-120b",
   "hosts": 22,
   "cheapest": {
    "host": "Near AI",
    "input": 0.03,
    "output": 0.15,
    "quantization": "fp4",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.06,
    "variants": 1
   },
   "dearest": {
    "host": "Cerebras",
    "input": 0.35,
    "output": 0.75,
    "quantization": "fp16",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.45,
    "variants": 1
   },
   "spreadRatio": 7.5,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Near AI",
     "input": 0.03,
     "output": 0.15,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.06,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.03,
     "output": 0.15,
     "quantization": "int4",
     "contextLength": 128000,
     "uptime30m": 100,
     "blended": 0.06,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.03,
     "output": 0.17,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.065,
     "variants": 1
    },
    {
     "host": "DekaLLM",
     "input": 0.03,
     "output": 0.18,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.0675,
     "variants": 1
    },
    {
     "host": "AkashML",
     "input": 0.03,
     "output": 0.187,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.06925,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.037,
     "output": 0.17,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.07025,
     "variants": 1
    },
    {
     "host": "Crusoe",
     "input": 0.05,
     "output": 0.25,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.1,
     "variants": 1
    },
    {
     "host": "Mancer 2",
     "input": 0.05,
     "output": 0.25,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.1,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.05,
     "output": 0.25,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.1,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.06,
     "output": 0.42,
     "quantization": "unknown",
     "contextLength": 128000,
     "uptime30m": 100,
     "blended": 0.15,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.09,
     "output": 0.36,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 0,
     "blended": 0.1575,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 0.1,
     "output": 0.5,
     "quantization": "fp4",
     "contextLength": 128072,
     "uptime30m": 100,
     "blended": 0.2,
     "variants": 2
    },
    {
     "host": "Parasail",
     "input": 0.1,
     "output": 0.75,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Amazon Bedrock",
     "input": 0.15,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 2
    },
    {
     "host": "Nebius",
     "input": 0.15,
     "output": 0.6,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.15,
     "output": 0.6,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.15,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 0.15,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Groq",
     "input": 0.15,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Mara",
     "input": 0.15,
     "output": 0.75,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.3,
     "variants": 1
    },
    {
     "host": "SambaNova",
     "input": 0.14,
     "output": 0.95,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.3425,
     "variants": 1
    },
    {
     "host": "Cerebras",
     "input": 0.35,
     "output": 0.75,
     "quantization": "fp16",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.45,
     "variants": 1
    }
   ]
  },
  {
   "model": "deepseek/deepseek-v4-flash",
   "huggingFaceId": "deepseek-ai/DeepSeek-V4-Flash",
   "hosts": 17,
   "cheapest": {
    "host": "StreamLake",
    "input": 0.0826,
    "output": 0.1652,
    "quantization": "fp8",
    "contextLength": 1024000,
    "uptime30m": 100,
    "blended": 0.10325,
    "variants": 1
   },
   "dearest": {
    "host": "Cloudflare",
    "input": 0.44,
    "output": 1.32,
    "quantization": "unknown",
    "contextLength": 384000,
    "uptime30m": 100,
    "blended": 0.66,
    "variants": 1
   },
   "spreadRatio": 6.39,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "StreamLake",
     "input": 0.0826,
     "output": 0.1652,
     "quantization": "fp8",
     "contextLength": 1024000,
     "uptime30m": 100,
     "blended": 0.10325,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.09,
     "output": 0.18,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.1125,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.091,
     "output": 0.182,
     "quantization": "fp8",
     "contextLength": 1048575,
     "uptime30m": 100,
     "blended": 0.11375,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.0966,
     "output": 0.1925,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.120575,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.098,
     "output": 0.196,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.1225,
     "variants": 1
    },
    {
     "host": "Fireworks",
     "input": 0.105,
     "output": 0.21,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 82.14285714285714,
     "blended": 0.13125,
     "variants": 1
    },
    {
     "host": "Relace",
     "input": 0.05,
     "output": 0.45,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.15,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.13,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.1675,
     "variants": 1
    },
    {
     "host": "Crusoe",
     "input": 0.14,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 0.14,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.14,
     "output": 0.28,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.14,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.15,
     "output": 0.3,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 90,
     "blended": 0.1875,
     "variants": 1
    },
    {
     "host": "Mancer 2",
     "input": 0.19,
     "output": 0.5,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2675,
     "variants": 1
    },
    {
     "host": "Azure",
     "input": 0.21,
     "output": 0.56,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 93.33333333333333,
     "blended": 0.2975,
     "variants": 1
    },
    {
     "host": "Wafer",
     "input": 0.6,
     "output": 0.17,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.4925,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 0.44,
     "output": 1.32,
     "quantization": "unknown",
     "contextLength": 384000,
     "uptime30m": 100,
     "blended": 0.66,
     "variants": 1
    }
   ]
  },
  {
   "model": "deepseek/deepseek-v4.1-flash",
   "huggingFaceId": "deepseek-ai/DeepSeek-V4.1-Flash",
   "hosts": 33,
   "cheapest": {
    "host": "Decart",
    "input": 0.09,
    "output": 0.18,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": 76.66666666666667,
    "blended": 0.1125,
    "variants": 1
   },
   "dearest": {
    "host": "Alibaba",
    "input": 0.33,
    "output": 1.32,
    "quantization": "unknown",
    "contextLength": 1000000,
    "uptime30m": 100,
    "blended": 0.5775,
    "variants": 1
   },
   "spreadRatio": 5.13,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Decart",
     "input": 0.09,
     "output": 0.18,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 76.66666666666667,
     "blended": 0.1125,
     "variants": 1
    },
    {
     "host": "Sail Research",
     "input": 0.09,
     "output": 0.25,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.13,
     "variants": 2
    },
    {
     "host": "DeepInfra",
     "input": 0.112,
     "output": 0.336,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.168,
     "variants": 3
    },
    {
     "host": "Relace",
     "input": 0.07,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.1775,
     "variants": 1
    },
    {
     "host": "Morph",
     "input": 0.1,
     "output": 0.6,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.225,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.135,
     "output": 0.54,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.23625,
     "variants": 1
    },
    {
     "host": "InferenceNet",
     "input": 0.103,
     "output": 0.645,
     "quantization": "fp8",
     "contextLength": 1040000,
     "uptime30m": 100,
     "blended": 0.2385,
     "variants": 2
    },
    {
     "host": "StreamLake",
     "input": 0.147,
     "output": 0.588,
     "quantization": "fp8",
     "contextLength": 1024000,
     "uptime30m": 100,
     "blended": 0.25725,
     "variants": 1
    },
    {
     "host": "OpenInference",
     "input": 0.02,
     "output": 1,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.265,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.15,
     "output": 0.65,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.275,
     "variants": 1
    },
    {
     "host": "Reka",
     "input": 0.105,
     "output": 0.84,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.28875,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.165,
     "output": 0.66,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.28875,
     "variants": 1
    },
    {
     "host": "ModelRun",
     "input": 0.17,
     "output": 0.67,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.295,
     "variants": 1
    },
    {
     "host": "Wafer",
     "input": 0.19,
     "output": 0.65,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.305,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.18,
     "output": 0.72,
     "quantization": "fp8",
     "contextLength": 1048575,
     "uptime30m": 100,
     "blended": 0.315,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.195,
     "output": 0.78,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.34125,
     "variants": 1
    },
    {
     "host": "Fireworks",
     "input": 0.21,
     "output": 0.84,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.3675,
     "variants": 3
    },
    {
     "host": "Phala",
     "input": 0.21,
     "output": 0.84,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.3675,
     "variants": 1
    },
    {
     "host": "Mancer 2",
     "input": 0.2,
     "output": 1,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 93.33333333333333,
     "blended": 0.4,
     "variants": 1
    },
    {
     "host": "Makora",
     "input": 0.26,
     "output": 1.17,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.4875,
     "variants": 1
    },
    {
     "host": "Crusoe",
     "input": 0.29,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.5175,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048000,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "PrimeIntellect",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Cohere",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 3
    },
    {
     "host": "Together",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Modal",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.33,
     "output": 1.32,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.5775,
     "variants": 1
    }
   ]
  },
  {
   "model": "deepseek/deepseek-v4.1-flash-1017",
   "huggingFaceId": "deepseek-ai/DeepSeek-V4.1-Flash-1017",
   "hosts": 30,
   "cheapest": {
    "host": "Sail Research",
    "input": 0.101,
    "output": 0.15,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.11325,
    "variants": 2
   },
   "dearest": {
    "host": "Alibaba",
    "input": 0.33,
    "output": 1.32,
    "quantization": "unknown",
    "contextLength": 1000000,
    "uptime30m": 100,
    "blended": 0.5775,
    "variants": 1
   },
   "spreadRatio": 5.1,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Sail Research",
     "input": 0.101,
     "output": 0.15,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.11325,
     "variants": 2
    },
    {
     "host": "DeepInfra",
     "input": 0.0784,
     "output": 0.2352,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.1176,
     "variants": 3
    },
    {
     "host": "Reka",
     "input": 0.04,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.155,
     "variants": 1
    },
    {
     "host": "Relace",
     "input": 0.07,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.1775,
     "variants": 1
    },
    {
     "host": "InferenceNet",
     "input": 0.08,
     "output": 0.5,
     "quantization": "fp8",
     "contextLength": 1040000,
     "uptime30m": 100,
     "blended": 0.185,
     "variants": 2
    },
    {
     "host": "Wafer",
     "input": 0.13,
     "output": 0.5,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2225,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 0.1,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.225,
     "variants": 1
    },
    {
     "host": "Morph",
     "input": 0.1,
     "output": 0.6,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.225,
     "variants": 1
    },
    {
     "host": "Decart",
     "input": 0.1365,
     "output": 0.546,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.238875,
     "variants": 1
    },
    {
     "host": "OpenInference",
     "input": 0.02,
     "output": 1,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.265,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.18,
     "output": 0.72,
     "quantization": "fp8",
     "contextLength": 1048575,
     "uptime30m": 100,
     "blended": 0.315,
     "variants": 1
    },
    {
     "host": "Ionstream",
     "input": 0.22,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.315,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.195,
     "output": 0.78,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.34125,
     "variants": 1
    },
    {
     "host": "Fireworks",
     "input": 0.21,
     "output": 0.84,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.3675,
     "variants": 2
    },
    {
     "host": "StreamLake",
     "input": 0.21,
     "output": 0.84,
     "quantization": "fp8",
     "contextLength": 1024000,
     "uptime30m": 100,
     "blended": 0.3675,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.225,
     "output": 0.9,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.39375,
     "variants": 1
    },
    {
     "host": "Makora",
     "input": 0.15,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.4125,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.24,
     "output": 0.96,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.42,
     "variants": 1
    },
    {
     "host": "Modal",
     "input": 0.25,
     "output": 1,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.4375,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 0.28,
     "output": 1.13,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.4925,
     "variants": 1
    },
    {
     "host": "Crusoe",
     "input": 0.29,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.5175,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "PrimeIntellect",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "DeepSeek",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.33,
     "output": 1.32,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.5775,
     "variants": 1
    }
   ]
  },
  {
   "model": "google/gemma-4-31b-it",
   "huggingFaceId": "google/gemma-4-31B-it",
   "hosts": 12,
   "cheapest": {
    "host": "CoreWeave",
    "input": 0.1,
    "output": 0.34,
    "quantization": "fp4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.16,
    "variants": 1
   },
   "dearest": {
    "host": "SiliconFlow",
    "input": 0.75,
    "output": 1,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.8125,
    "variants": 1
   },
   "spreadRatio": 5.08,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "CoreWeave",
     "input": 0.1,
     "output": 0.34,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.16,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.12,
     "output": 0.36,
     "quantization": "fp4",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 0.18,
     "variants": 1
    },
    {
     "host": "Chutes",
     "input": 0.12,
     "output": 0.37,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.1825,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.16,
     "output": 0.32,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.2,
     "variants": 4
    },
    {
     "host": "Crusoe",
     "input": 0.14,
     "output": 0.4,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.205,
     "variants": 1
    },
    {
     "host": "Friendli",
     "input": 0.14,
     "output": 0.4,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.205,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.14,
     "output": 0.4,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.205,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.15,
     "output": 0.4,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.2125,
     "variants": 1
    },
    {
     "host": "Io Net",
     "input": 0.361,
     "output": 1.0925,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.543875,
     "variants": 1
    },
    {
     "host": "SambaNova",
     "input": 0.38,
     "output": 1.15,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.5725,
     "variants": 1
    },
    {
     "host": "ModelRun",
     "input": 0.75,
     "output": 1,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.8125,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.75,
     "output": 1,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.8125,
     "variants": 1
    }
   ]
  },
  {
   "model": "meta-llama/llama-3.3-70b-instruct",
   "huggingFaceId": "meta-llama/Llama-3.3-70B-Instruct",
   "hosts": 9,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.1,
    "output": 0.32,
    "quantization": "fp8",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.155,
    "variants": 1
   },
   "dearest": {
    "host": "Cloudflare",
    "input": 0.293,
    "output": 2.253,
    "quantization": "fp8",
    "contextLength": 24000,
    "uptime30m": 100,
    "blended": 0.783,
    "variants": 1
   },
   "spreadRatio": 5.05,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.1,
     "output": 0.32,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.155,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.135,
     "output": 0.4,
     "quantization": "bf16",
     "contextLength": 12288,
     "uptime30m": 100,
     "blended": 0.20125,
     "variants": 1
    },
    {
     "host": "AkashML",
     "input": 0.2,
     "output": 0.52,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.28,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.22,
     "output": 0.5,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.29,
     "variants": 1
    },
    {
     "host": "SambaNova",
     "input": 0.45,
     "output": 0.9,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.5625,
     "variants": 1
    },
    {
     "host": "Groq",
     "input": 0.59,
     "output": 0.79,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.64,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.71,
     "output": 0.71,
     "quantization": "fp16",
     "contextLength": 128000,
     "uptime30m": 100,
     "blended": 0.71,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.72,
     "output": 0.72,
     "quantization": "unknown",
     "contextLength": 128000,
     "uptime30m": null,
     "blended": 0.72,
     "variants": 2
    },
    {
     "host": "Cloudflare",
     "input": 0.293,
     "output": 2.253,
     "quantization": "fp8",
     "contextLength": 24000,
     "uptime30m": 100,
     "blended": 0.783,
     "variants": 1
    }
   ]
  },
  {
   "model": "nvidia/nemotron-3-super-120b-a12b",
   "huggingFaceId": "nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-FP8",
   "hosts": 3,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.085,
    "output": 0.4,
    "quantization": "bf16",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.16375,
    "variants": 1
   },
   "dearest": {
    "host": "Crusoe",
    "input": 0.3,
    "output": 2.4,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.825,
    "variants": 1
   },
   "spreadRatio": 5.04,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.085,
     "output": 0.4,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.16375,
     "variants": 1
    },
    {
     "host": "DekaLLM",
     "input": 0.08,
     "output": 0.45,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1725,
     "variants": 1
    },
    {
     "host": "Crusoe",
     "input": 0.3,
     "output": 2.4,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.825,
     "variants": 1
    }
   ]
  },
  {
   "model": "gryphe/mythomax-l2-13b",
   "huggingFaceId": "Gryphe/MythoMax-L2-13b",
   "hosts": 3,
   "cheapest": {
    "host": "Parasail",
    "input": 0.08,
    "output": 0.11,
    "quantization": "fp16",
    "contextLength": 4096,
    "uptime30m": 100,
    "blended": 0.0875,
    "variants": 1
   },
   "dearest": {
    "host": "Mancer 2",
    "input": 0.35,
    "output": 0.6,
    "quantization": "fp8",
    "contextLength": 8192,
    "uptime30m": 100,
    "blended": 0.4125,
    "variants": 1
   },
   "spreadRatio": 4.71,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Parasail",
     "input": 0.08,
     "output": 0.11,
     "quantization": "fp16",
     "contextLength": 4096,
     "uptime30m": 100,
     "blended": 0.0875,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.4,
     "output": 0.4,
     "quantization": "fp16",
     "contextLength": 4096,
     "uptime30m": null,
     "blended": 0.4,
     "variants": 1
    },
    {
     "host": "Mancer 2",
     "input": 0.35,
     "output": 0.6,
     "quantization": "fp8",
     "contextLength": 8192,
     "uptime30m": 100,
     "blended": 0.4125,
     "variants": 1
    }
   ]
  },
  {
   "model": "deepseek/deepseek-v4-pro-0813",
   "huggingFaceId": "deepseek-ai/DeepSeek-V4-Pro-0813",
   "hosts": 22,
   "cheapest": {
    "host": "Sail Research",
    "input": 0.45,
    "output": 0.8,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.5375,
    "variants": 2
   },
   "dearest": {
    "host": "Venice",
    "input": 1.65,
    "output": 4.95,
    "quantization": "unknown",
    "contextLength": 1000000,
    "uptime30m": null,
    "blended": 2.475,
    "variants": 1
   },
   "spreadRatio": 4.6,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Sail Research",
     "input": 0.45,
     "output": 0.8,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.5375,
     "variants": 2
    },
    {
     "host": "StreamLake",
     "input": 0.66,
     "output": 1.98,
     "quantization": "unknown",
     "contextLength": 1024000,
     "uptime30m": 100,
     "blended": 0.99,
     "variants": 1
    },
    {
     "host": "Ionstream",
     "input": 0.37,
     "output": 2.93,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.01,
     "variants": 1
    },
    {
     "host": "Wafer",
     "input": 0.285,
     "output": 4.3,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.28875,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.99,
     "output": 2.97,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.485,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 1.056,
     "output": 3.168,
     "quantization": "fp8",
     "contextLength": 1048575,
     "uptime30m": null,
     "blended": 1.584,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 1.3,
     "output": 2.6,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.625,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 1.122,
     "output": 3.366,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 1.683,
     "variants": 2
    },
    {
     "host": "DekaLLM",
     "input": 1.08,
     "output": 3.8,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.76,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 1.31,
     "output": 3.96,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.9725,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 1.32,
     "output": 3.96,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.98,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 1.32,
     "output": 3.96,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.98,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 1.32,
     "output": 3.96,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.98,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 1.32,
     "output": 3.96,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.98,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 1.32,
     "output": 3.96,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.98,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 1.32,
     "output": 3.96,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.98,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 1.32,
     "output": 3.96,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.98,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 1.32,
     "output": 3.96,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.98,
     "variants": 1
    },
    {
     "host": "DeepSeek",
     "input": 1.32,
     "output": 3.96,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.98,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 1.45,
     "output": 4.36,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 2.1775,
     "variants": 1
    },
    {
     "host": "Relace",
     "input": 0.2565,
     "output": 8,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.192375,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 1.65,
     "output": 4.95,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": null,
     "blended": 2.475,
     "variants": 1
    }
   ]
  },
  {
   "model": "z-ai/glm-5.2",
   "huggingFaceId": "zai-org/GLM-5.2",
   "hosts": 25,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.45,
    "output": 1.44,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": null,
    "blended": 0.6975,
    "variants": 2
   },
   "dearest": {
    "host": "Decart",
    "input": 1.2,
    "output": 8,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 2.9,
    "variants": 1
   },
   "spreadRatio": 4.16,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.45,
     "output": 1.44,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 0.6975,
     "variants": 2
    },
    {
     "host": "StreamLake",
     "input": 0.61306,
     "output": 1.92676,
     "quantization": "fp8",
     "contextLength": 1024000,
     "uptime30m": 100,
     "blended": 0.941485,
     "variants": 1
    },
    {
     "host": "Ambient",
     "input": 0.6,
     "output": 2,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 0.95,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.6496,
     "output": 2.0416,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.9976,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.62,
     "output": 2.42,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.07,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.7,
     "output": 2.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.075,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.7,
     "output": 2.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.075,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.938,
     "output": 2.948,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.4405,
     "variants": 1
    },
    {
     "host": "Inceptron",
     "input": 0.99,
     "output": 2.88,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.4625,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.966,
     "output": 3.036,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 1.4835,
     "variants": 3
    },
    {
     "host": "Morph",
     "input": 0.669,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.60175,
     "variants": 1
    },
    {
     "host": "Wafer",
     "input": 0.0341,
     "output": 7,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.775575,
     "variants": 2
    },
    {
     "host": "Relace",
     "input": 0.0324,
     "output": 7.1524,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.8124,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 0.19,
     "output": 8,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 2.1425,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Mistral",
     "input": 1.4,
     "output": 4.4,
     "quantization": "nvfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 2
    },
    {
     "host": "BaseTen",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 4
    },
    {
     "host": "Together",
     "input": 1.4,
     "output": 4.4,
     "quantization": "unknown",
     "contextLength": 1048575,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 2
    },
    {
     "host": "Venice",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Friendli",
     "input": 1.4,
     "output": 4.4,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Z.AI",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Decart",
     "input": 1.2,
     "output": 8,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.9,
     "variants": 1
    }
   ]
  },
  {
   "model": "nvidia/nemotron-3.5-super-vl",
   "huggingFaceId": "nvidia/NVIDIA-Nemotron-3.5-Super-VL",
   "hosts": 8,
   "cheapest": {
    "host": "Darkbloom",
    "input": 0.06,
    "output": 0.35,
    "quantization": "nvfp4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.1325,
    "variants": 1
   },
   "dearest": {
    "host": "Venice",
    "input": 0.25,
    "output": 1.25,
    "quantization": "fp4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.5,
    "variants": 1
   },
   "spreadRatio": 3.77,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Darkbloom",
     "input": 0.06,
     "output": 0.35,
     "quantization": "nvfp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1325,
     "variants": 1
    },
    {
     "host": "Crusoe",
     "input": 0.1,
     "output": 0.4,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "Io Net",
     "input": 0.15,
     "output": 0.5,
     "quantization": "nvfp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 0.15,
     "output": 0.5,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.15,
     "output": 0.5,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.15,
     "output": 0.5,
     "quantization": "nvfp4",
     "contextLength": 1048576,
     "uptime30m": 91.66666666666666,
     "blended": 0.2375,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 0.2,
     "output": 0.5,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.275,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.25,
     "output": 1.25,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.5,
     "variants": 1
    }
   ]
  },
  {
   "model": "openai/gpt-oss-20b",
   "huggingFaceId": "openai/gpt-oss-20b",
   "hosts": 10,
   "cheapest": {
    "host": "Darkbloom",
    "input": 0.018,
    "output": 0.09,
    "quantization": "fp8",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.036,
    "variants": 1
   },
   "dearest": {
    "host": "Groq",
    "input": 0.075,
    "output": 0.3,
    "quantization": "unknown",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.13125,
    "variants": 1
   },
   "spreadRatio": 3.65,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Darkbloom",
     "input": 0.018,
     "output": 0.09,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.036,
     "variants": 1
    },
    {
     "host": "AkashML",
     "input": 0.02,
     "output": 0.1,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.04,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.03,
     "output": 0.13,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.055,
     "variants": 1
    },
    {
     "host": "DekaLLM",
     "input": 0.029,
     "output": 0.14,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.05675,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.03,
     "output": 0.14,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.0575,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.03,
     "output": 0.15,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.06,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.04,
     "output": 0.18,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 93.33333333333333,
     "blended": 0.075,
     "variants": 1
    },
    {
     "host": "Amazon Bedrock",
     "input": 0.07,
     "output": 0.15,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": null,
     "blended": 0.09,
     "variants": 2
    },
    {
     "host": "Google",
     "input": 0.07,
     "output": 0.25,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": null,
     "blended": 0.115,
     "variants": 1
    },
    {
     "host": "Groq",
     "input": 0.075,
     "output": 0.3,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.13125,
     "variants": 1
    }
   ]
  },
  {
   "model": "moonshotai/kimi-k3",
   "huggingFaceId": "moonshotai/Kimi-K3",
   "hosts": 21,
   "cheapest": {
    "host": "Sail Research",
    "input": 1.2,
    "output": 5.5,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 2.275,
    "variants": 2
   },
   "dearest": {
    "host": "Alibaba",
    "input": 3.45,
    "output": 17.25,
    "quantization": "unknown",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 6.9,
    "variants": 1
   },
   "spreadRatio": 3.03,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Sail Research",
     "input": 1.2,
     "output": 5.5,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.275,
     "variants": 2
    },
    {
     "host": "Relace",
     "input": 0.2033,
     "output": 8.5226,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 92.5925925925926,
     "blended": 2.283125,
     "variants": 1
    },
    {
     "host": "InferenceNet",
     "input": 0.239,
     "output": 10.8,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.87925,
     "variants": 4
    },
    {
     "host": "Wafer",
     "input": 0.36,
     "output": 11.9,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 3.245,
     "variants": 2
    },
    {
     "host": "Decart",
     "input": 1.65,
     "output": 8.25,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 3.3,
     "variants": 1
    },
    {
     "host": "Chutes",
     "input": 1.85,
     "output": 9.5,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 3.7625,
     "variants": 1
    },
    {
     "host": "Fireworks",
     "input": 2.1,
     "output": 10.5,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 4.2,
     "variants": 4
    },
    {
     "host": "Makora",
     "input": 0.65,
     "output": 15,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 96.55172413793103,
     "blended": 4.2375,
     "variants": 1
    },
    {
     "host": "AkashML",
     "input": 1.2,
     "output": 14,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 4.4,
     "variants": 1
    },
    {
     "host": "Morph",
     "input": 0.9,
     "output": 17.379,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 5.01975,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 2.55,
     "output": 12.75,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 5.1,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 2.55,
     "output": 12.95,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 5.15,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 2.85,
     "output": 14.25,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 5.7,
     "variants": 1
    },
    {
     "host": "ParetoAlpha",
     "input": 3,
     "output": 15,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 6,
     "variants": 1
    },
    {
     "host": "Amazon Bedrock",
     "input": 3,
     "output": 15,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 6,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 3,
     "output": 15,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 91.66666666666666,
     "blended": 6,
     "variants": 1
    },
    {
     "host": "Modal",
     "input": 3,
     "output": 15,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 6,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 3,
     "output": 15,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 44.44444444444444,
     "blended": 6,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 3,
     "output": 15,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 6,
     "variants": 1
    },
    {
     "host": "Moonshot AI",
     "input": 3,
     "output": 15,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 6,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 3.45,
     "output": 17.25,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 6.9,
     "variants": 1
    }
   ]
  },
  {
   "model": "google/gemma-4-26b-a4b-it",
   "huggingFaceId": "google/gemma-4-26B-A4B-it",
   "hosts": 13,
   "cheapest": {
    "host": "Darkbloom",
    "input": 0.042,
    "output": 0.22,
    "quantization": "unknown",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.0865,
    "variants": 1
   },
   "dearest": {
    "host": "Google",
    "input": 0.15,
    "output": 0.6,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.2625,
    "variants": 1
   },
   "spreadRatio": 3.03,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Darkbloom",
     "input": 0.042,
     "output": 0.22,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.0865,
     "variants": 1
    },
    {
     "host": "Makora",
     "input": 0.07,
     "output": 0.24,
     "quantization": "unknown",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 0.1125,
     "variants": 1
    },
    {
     "host": "DekaLLM",
     "input": 0.06,
     "output": 0.33,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1275,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.07,
     "output": 0.34,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1375,
     "variants": 1
    },
    {
     "host": "NextBit",
     "input": 0.09,
     "output": 0.3,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1425,
     "variants": 1
    },
    {
     "host": "Io Net",
     "input": 0.09,
     "output": 0.31,
     "quantization": "bf16",
     "contextLength": 262142,
     "uptime30m": 100,
     "blended": 0.145,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.1,
     "output": 0.3,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.15,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 0.1,
     "output": 0.3,
     "quantization": "unknown",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 0.15,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.13,
     "output": 0.4,
     "quantization": "bf16",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 0.1975,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.13,
     "output": 0.4,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1975,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.13,
     "output": 0.4,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1975,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.14,
     "output": 0.4,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 14.285714285714285,
     "blended": 0.205,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.15,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3.6-35b-a3b",
   "huggingFaceId": "Qwen/Qwen3.6-35B-A3B",
   "hosts": 10,
   "cheapest": {
    "host": "Darkbloom",
    "input": 0.05,
    "output": 0.7,
    "quantization": "fp4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.2125,
    "variants": 1
   },
   "dearest": {
    "host": "SiliconFlow",
    "input": 0.24,
    "output": 1.8,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": null,
    "blended": 0.63,
    "variants": 1
   },
   "spreadRatio": 2.96,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Darkbloom",
     "input": 0.05,
     "output": 0.7,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.2125,
     "variants": 1
    },
    {
     "host": "AkashML",
     "input": 0.1,
     "output": 0.9,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.3,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.1,
     "output": 0.95,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.3125,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.1,
     "output": 0.95,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.3125,
     "variants": 1
    },
    {
     "host": "DekaLLM",
     "input": 0.1,
     "output": 1,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.325,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.1,
     "output": 1,
     "quantization": "fp8",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 0.325,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.2,
     "output": 1.27,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.4675,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.25,
     "output": 1.25,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.5,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.25,
     "output": 1.25,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.5,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.24,
     "output": 1.8,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 0.63,
     "variants": 1
    }
   ]
  },
  {
   "model": "minimax/minimax-m3",
   "huggingFaceId": "MiniMaxAI/Minimax-M3",
   "hosts": 12,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.224,
    "output": 0.88,
    "quantization": "fp8",
    "contextLength": 524288,
    "uptime30m": null,
    "blended": 0.388,
    "variants": 2
   },
   "dearest": {
    "host": "Mara",
    "input": 0.6,
    "output": 2.4,
    "quantization": "unknown",
    "contextLength": 1048576,
    "uptime30m": null,
    "blended": 1.05,
    "variants": 1
   },
   "spreadRatio": 2.71,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.224,
     "output": 0.88,
     "quantization": "fp8",
     "contextLength": 524288,
     "uptime30m": null,
     "blended": 0.388,
     "variants": 2
    },
    {
     "host": "CoreWeave",
     "input": 0.23,
     "output": 0.96,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.4125,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.24,
     "output": 0.96,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.42,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 524288,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 524288,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 524300,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Minimax",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 524288,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "ModelRun",
     "input": 0.3,
     "output": 1.3,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.55,
     "variants": 1
    },
    {
     "host": "Mara",
     "input": 0.6,
     "output": 2.4,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.05,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3.5-35b-a3b",
   "huggingFaceId": "Qwen/Qwen3.5-35B-A3B",
   "hosts": 7,
   "cheapest": {
    "host": "Darkbloom",
    "input": 0.065,
    "output": 0.75,
    "quantization": "fp4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.23625,
    "variants": 1
   },
   "dearest": {
    "host": "SiliconFlow",
    "input": 0.24,
    "output": 1.8,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": null,
    "blended": 0.63,
    "variants": 1
   },
   "spreadRatio": 2.67,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Darkbloom",
     "input": 0.065,
     "output": 0.75,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.23625,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.14,
     "output": 1,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.355,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.15,
     "output": 1,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.3625,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.15,
     "output": 1,
     "quantization": "unknown",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 0.3625,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.1625,
     "output": 1.3,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.446875,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.225,
     "output": 1.8,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 0.61875,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.24,
     "output": 1.8,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 0.63,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3-235b-a22b-2507",
   "huggingFaceId": "Qwen/Qwen3-235B-A22B-Instruct-2507",
   "hosts": 5,
   "cheapest": {
    "host": "GMICloud",
    "input": 0.0875,
    "output": 0.35,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 0,
    "blended": 0.153125,
    "variants": 1
   },
   "dearest": {
    "host": "Google",
    "input": 0.22,
    "output": 0.88,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 0,
    "blended": 0.385,
    "variants": 2
   },
   "spreadRatio": 2.51,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "GMICloud",
     "input": 0.0875,
     "output": 0.35,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 0,
     "blended": 0.153125,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.09,
     "output": 0.55,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.205,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.15,
     "output": 0.75,
     "quantization": "fp8",
     "contextLength": 128000,
     "uptime30m": 100,
     "blended": 0.3,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.14,
     "output": 0.8,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.305,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.22,
     "output": 0.88,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 0,
     "blended": 0.385,
     "variants": 2
    }
   ]
  },
  {
   "model": "minimax/minimax-m3.1-flash",
   "huggingFaceId": "MiniMaxAI/MiniMax-M3.1-Flash",
   "hosts": 11,
   "cheapest": {
    "host": "InferenceNet",
    "input": 0.1,
    "output": 0.55,
    "quantization": "fp4",
    "contextLength": 1000000,
    "uptime30m": 100,
    "blended": 0.2125,
    "variants": 2
   },
   "dearest": {
    "host": "SambaNova",
    "input": 0.3,
    "output": 1.2,
    "quantization": "unknown",
    "contextLength": 1048576,
    "uptime30m": null,
    "blended": 0.525,
    "variants": 1
   },
   "spreadRatio": 2.47,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "InferenceNet",
     "input": 0.1,
     "output": 0.55,
     "quantization": "fp4",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.2125,
     "variants": 2
    },
    {
     "host": "GMICloud",
     "input": 0.1425,
     "output": 0.57,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.249375,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.15,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.15,
     "output": 0.6,
     "quantization": "fp16",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 0.15,
     "output": 0.6,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Wafer",
     "input": 0.15,
     "output": 0.6,
     "quantization": "nvfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "ModelRun",
     "input": 0.15,
     "output": 0.6,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.15,
     "output": 0.6,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 0.15,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Minimax",
     "input": 0.15,
     "output": 0.6,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "SambaNova",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 0.525,
     "variants": 1
    }
   ]
  },
  {
   "model": "minimax/minimax-m2.7",
   "huggingFaceId": "MiniMaxAI/MiniMax-M2.7",
   "hosts": 5,
   "cheapest": {
    "host": "GMICloud",
    "input": 0.21,
    "output": 0.84,
    "quantization": "fp8",
    "contextLength": 196608,
    "uptime30m": 100,
    "blended": 0.3675,
    "variants": 1
   },
   "dearest": {
    "host": "Groq",
    "input": 0.6,
    "output": 1.8,
    "quantization": "unknown",
    "contextLength": 196608,
    "uptime30m": 66.66666666666666,
    "blended": 0.9,
    "variants": 1
   },
   "spreadRatio": 2.45,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "GMICloud",
     "input": 0.21,
     "output": 0.84,
     "quantization": "fp8",
     "contextLength": 196608,
     "uptime30m": 100,
     "blended": 0.3675,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.27,
     "output": 1.08,
     "quantization": "fp8",
     "contextLength": 204800,
     "uptime30m": 96.42857142857143,
     "blended": 0.4725,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 196608,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Minimax",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 204800,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 2
    },
    {
     "host": "Groq",
     "input": 0.6,
     "output": 1.8,
     "quantization": "unknown",
     "contextLength": 196608,
     "uptime30m": 66.66666666666666,
     "blended": 0.9,
     "variants": 1
    }
   ]
  },
  {
   "model": "deepseek/deepseek-v4-pro",
   "huggingFaceId": "deepseek-ai/DeepSeek-V4-Pro",
   "hosts": 15,
   "cheapest": {
    "host": "Reka",
    "input": 0.8,
    "output": 1.6,
    "quantization": "unknown",
    "contextLength": 1048576,
    "uptime30m": 86.66666666666667,
    "blended": 1,
    "variants": 1
   },
   "dearest": {
    "host": "Azure",
    "input": 1.91,
    "output": 3.83,
    "quantization": "unknown",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 2.39,
    "variants": 1
   },
   "spreadRatio": 2.39,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Reka",
     "input": 0.8,
     "output": 1.6,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 86.66666666666667,
     "blended": 1,
     "variants": 1
    },
    {
     "host": "Relace",
     "input": 0.9,
     "output": 1.5,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.05,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.95526,
     "output": 1.91052,
     "quantization": "fp8",
     "contextLength": 1024000,
     "uptime30m": 100,
     "blended": 1.194075,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.957,
     "output": 1.914,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.19625,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 1.044,
     "output": 2.088,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.305,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 1.15,
     "output": 2.55,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 93.33333333333333,
     "blended": 1.5,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 1.3,
     "output": 2.6,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.625,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 1.416,
     "output": 2.832,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 1.77,
     "variants": 2
    },
    {
     "host": "SiliconFlow",
     "input": 1.50162,
     "output": 3.135,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 96.66666666666667,
     "blended": 1.909965,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 1.6,
     "output": 3.2,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 1.65,
     "output": 3.301,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 2.06275,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 1.68,
     "output": 3.38,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.105,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 1.69,
     "output": 3.38,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.1125,
     "variants": 1
    },
    {
     "host": "Crusoe",
     "input": 1.74,
     "output": 3.48,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 53.84615384615385,
     "blended": 2.175,
     "variants": 1
    },
    {
     "host": "Azure",
     "input": 1.91,
     "output": 3.83,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2.39,
     "variants": 1
    }
   ]
  },
  {
   "model": "moonshotai/kimi-k2.6",
   "huggingFaceId": "moonshotai/Kimi-K2.6",
   "hosts": 17,
   "cheapest": {
    "host": "Chutes",
    "input": 0.45,
    "output": 1.95,
    "quantization": "int4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.825,
    "variants": 1
   },
   "dearest": {
    "host": "Phala",
    "input": 1.09,
    "output": 4.6,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 1.9675,
    "variants": 1
   },
   "spreadRatio": 2.38,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Chutes",
     "input": 0.45,
     "output": 1.95,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.825,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.57,
     "output": 2.4,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.0275,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.5985,
     "output": 2.52,
     "quantization": "fp8",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 1.078875,
     "variants": 1
    },
    {
     "host": "Inceptron",
     "input": 0.603,
     "output": 3.2,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.25225,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.65,
     "output": 3.41,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.34,
     "variants": 1
    },
    {
     "host": "Crusoe",
     "input": 0.7,
     "output": 3.5,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 93.10344827586206,
     "blended": 1.4,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.77,
     "output": 3.4,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.4275,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.75,
     "output": 3.5,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.4375,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.75,
     "output": 3.5,
     "quantization": "int4",
     "contextLength": 256000,
     "uptime30m": 85.71428571428571,
     "blended": 1.4375,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.75,
     "output": 3.5,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.4375,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.8,
     "output": 3.4,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.45,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.855,
     "output": 3.6,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 0,
     "blended": 1.54125,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 0.95,
     "output": 4,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.7125,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.95,
     "output": 4,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.7125,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 0.95,
     "output": 4,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.7125,
     "variants": 1
    },
    {
     "host": "Moonshot AI",
     "input": 0.95,
     "output": 4,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.7125,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 1.09,
     "output": 4.6,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.9675,
     "variants": 1
    }
   ]
  },
  {
   "model": "xiaomi/mimo-v2.5-pro",
   "huggingFaceId": "XiaomiMiMo/MiMo-V2.5-Pro",
   "hosts": 3,
   "cheapest": {
    "host": "GMICloud",
    "input": 0.3045,
    "output": 0.609,
    "quantization": "bf16",
    "contextLength": 1050000,
    "uptime30m": 0,
    "blended": 0.380625,
    "variants": 1
   },
   "dearest": {
    "host": "DigitalOcean",
    "input": 0.48,
    "output": 1.8,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.81,
    "variants": 1
   },
   "spreadRatio": 2.13,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "GMICloud",
     "input": 0.3045,
     "output": 0.609,
     "quantization": "bf16",
     "contextLength": 1050000,
     "uptime30m": 0,
     "blended": 0.380625,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.435,
     "output": 0.87,
     "quantization": "fp8",
     "contextLength": 1024000,
     "uptime30m": 100,
     "blended": 0.54375,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.48,
     "output": 1.8,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.81,
     "variants": 1
    }
   ]
  },
  {
   "model": "nvidia/nemotron-3.5-lightning",
   "huggingFaceId": "nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16",
   "hosts": 4,
   "cheapest": {
    "host": "Darkbloom",
    "input": 0.025,
    "output": 0.12,
    "quantization": "int4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.04875,
    "variants": 1
   },
   "dearest": {
    "host": "CoreWeave",
    "input": 0.07,
    "output": 0.2,
    "quantization": "bf16",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.1025,
    "variants": 1
   },
   "spreadRatio": 2.1,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Darkbloom",
     "input": 0.025,
     "output": 0.12,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.04875,
     "variants": 1
    },
    {
     "host": "Io Net",
     "input": 0.0469,
     "output": 0.134,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 0.068675,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.06,
     "output": 0.16,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.085,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.07,
     "output": 0.2,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1025,
     "variants": 1
    }
   ]
  },
  {
   "model": "deepseek/deepseek-chat-v3.1",
   "huggingFaceId": "deepseek-ai/DeepSeek-V3.1",
   "hosts": 5,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.25,
    "output": 0.95,
    "quantization": "fp4",
    "contextLength": 163840,
    "uptime30m": 100,
    "blended": 0.425,
    "variants": 1
   },
   "dearest": {
    "host": "Mara",
    "input": 0.6,
    "output": 1.7,
    "quantization": "unknown",
    "contextLength": 131072,
    "uptime30m": 0,
    "blended": 0.875,
    "variants": 1
   },
   "spreadRatio": 2.06,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.25,
     "output": 0.95,
     "quantization": "fp4",
     "contextLength": 163840,
     "uptime30m": 100,
     "blended": 0.425,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.27,
     "output": 1,
     "quantization": "fp8",
     "contextLength": 163840,
     "uptime30m": 93.33333333333333,
     "blended": 0.4525,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.55,
     "output": 1.65,
     "quantization": "fp8",
     "contextLength": 161000,
     "uptime30m": 100,
     "blended": 0.825,
     "variants": 1
    },
    {
     "host": "SambaNova",
     "input": 0.65,
     "output": 1.5,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 72.41379310344827,
     "blended": 0.8625,
     "variants": 1
    },
    {
     "host": "Mara",
     "input": 0.6,
     "output": 1.7,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 0,
     "blended": 0.875,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3.5-9b",
   "huggingFaceId": "Qwen/Qwen3.5-9B",
   "hosts": 6,
   "cheapest": {
    "host": "Darkbloom",
    "input": 0.08,
    "output": 0.13,
    "quantization": "fp4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.0925,
    "variants": 1
   },
   "dearest": {
    "host": "Together",
    "input": 0.17,
    "output": 0.25,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.19,
    "variants": 1
   },
   "spreadRatio": 2.05,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Darkbloom",
     "input": 0.08,
     "output": 0.13,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.0925,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.1,
     "output": 0.15,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1125,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.1,
     "output": 0.15,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 84.61538461538461,
     "blended": 0.1125,
     "variants": 1
    },
    {
     "host": "PrimeIntellect",
     "input": 0.1,
     "output": 0.25,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1375,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.1,
     "output": 0.25,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1375,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 0.17,
     "output": 0.25,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.19,
     "variants": 1
    }
   ]
  },
  {
   "model": "deepseek/deepseek-v4-flash-vision-exp",
   "huggingFaceId": "deepseek-ai/DeepSeek-V4-Flash-Vision-Exp",
   "hosts": 3,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.2156,
    "output": 0.6468,
    "quantization": "fp8",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.3234,
    "variants": 1
   },
   "dearest": {
    "host": "SiliconFlow",
    "input": 0.44,
    "output": 1.32,
    "quantization": "fp8",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.66,
    "variants": 1
   },
   "spreadRatio": 2.04,
   "sameQuantisationEverywhere": true,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.2156,
     "output": 0.6468,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.3234,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.44,
     "output": 1.32,
     "quantization": "fp8",
     "contextLength": 1048575,
     "uptime30m": 100,
     "blended": 0.66,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.44,
     "output": 1.32,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.66,
     "variants": 1
    }
   ]
  },
  {
   "model": "z-ai/glm-4.7",
   "huggingFaceId": "zai-org/GLM-4.7",
   "hosts": 6,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.32,
    "output": 1.4,
    "quantization": "fp4",
    "contextLength": 202752,
    "uptime30m": null,
    "blended": 0.59,
    "variants": 2
   },
   "dearest": {
    "host": "Mancer 2",
    "input": 0.7,
    "output": 2.5,
    "quantization": "fp4",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 1.15,
    "variants": 1
   },
   "spreadRatio": 1.95,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.32,
     "output": 1.4,
     "quantization": "fp4",
     "contextLength": 202752,
     "uptime30m": null,
     "blended": 0.59,
     "variants": 2
    },
    {
     "host": "Venice",
     "input": 0.4004,
     "output": 1.9292,
     "quantization": "fp4",
     "contextLength": 198000,
     "uptime30m": 100,
     "blended": 0.7826,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.54,
     "output": 1.98,
     "quantization": "fp8",
     "contextLength": 204800,
     "uptime30m": 100,
     "blended": 0.9,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.6,
     "output": 2.2,
     "quantization": "unknown",
     "contextLength": 200000,
     "uptime30m": 0,
     "blended": 1,
     "variants": 1
    },
    {
     "host": "Z.AI",
     "input": 0.6,
     "output": 2.2,
     "quantization": "fp4",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 1,
     "variants": 1
    },
    {
     "host": "Mancer 2",
     "input": 0.7,
     "output": 2.5,
     "quantization": "fp4",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 1.15,
     "variants": 1
    }
   ]
  },
  {
   "model": "xiaomi/mimo-v2.6-flash",
   "huggingFaceId": "XiaomiMiMo/MiMo-V2.6-Flash-RL",
   "hosts": 10,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.091,
    "output": 0.182,
    "quantization": "fp8",
    "contextLength": 1048576,
    "uptime30m": 56.666666666666664,
    "blended": 0.11375,
    "variants": 1
   },
   "dearest": {
    "host": "Venice",
    "input": 0.175,
    "output": 0.35,
    "quantization": "fp8",
    "contextLength": 1000000,
    "uptime30m": 0,
    "blended": 0.21875,
    "variants": 1
   },
   "spreadRatio": 1.92,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.091,
     "output": 0.182,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 56.666666666666664,
     "blended": 0.11375,
     "variants": 1
    },
    {
     "host": "Io Net",
     "input": 0.1,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048320,
     "uptime30m": 100,
     "blended": 0.145,
     "variants": 1
    },
    {
     "host": "Darkbloom",
     "input": 0.1,
     "output": 0.28,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.145,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.119,
     "output": 0.238,
     "quantization": "bf16",
     "contextLength": 1050000,
     "uptime30m": 100,
     "blended": 0.14875,
     "variants": 1
    },
    {
     "host": "Makora",
     "input": 0.13,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.1675,
     "variants": 1
    },
    {
     "host": "Relace",
     "input": 0.13,
     "output": 0.28,
     "quantization": "mxfp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.1675,
     "variants": 1
    },
    {
     "host": "Cohere",
     "input": 0.14,
     "output": 0.28,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 93.33333333333333,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.14,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "Xiaomi",
     "input": 0.14,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.175,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.175,
     "output": 0.35,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 0,
     "blended": 0.21875,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3.5-397b-a17b",
   "huggingFaceId": "Qwen/Qwen3.5-397B-A17B",
   "hosts": 10,
   "cheapest": {
    "host": "Alibaba",
    "input": 0.39,
    "output": 2.34,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.8775,
    "variants": 1
   },
   "dearest": {
    "host": "Venice",
    "input": 0.75,
    "output": 4.5,
    "quantization": "unknown",
    "contextLength": 128000,
    "uptime30m": null,
    "blended": 1.6875,
    "variants": 1
   },
   "spreadRatio": 1.92,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Alibaba",
     "input": 0.39,
     "output": 2.34,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.8775,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.45,
     "output": 3,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.0875,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.5,
     "output": 3.6,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.275,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.55,
     "output": 3.5,
     "quantization": "unknown",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 1.2875,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.55,
     "output": 3.5,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.2875,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.55,
     "output": 3.5,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 1.2875,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.6,
     "output": 3.6,
     "quantization": "unknown",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 1.35,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.6,
     "output": 3.6,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 1.35,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.6,
     "output": 3.6,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.35,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.75,
     "output": 4.5,
     "quantization": "unknown",
     "contextLength": 128000,
     "uptime30m": null,
     "blended": 1.6875,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3-32b",
   "huggingFaceId": "Qwen/Qwen3-32B",
   "hosts": 3,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.08,
    "output": 0.28,
    "quantization": "fp8",
    "contextLength": 40960,
    "uptime30m": 100,
    "blended": 0.13,
    "variants": 1
   },
   "dearest": {
    "host": "SiliconFlow",
    "input": 0.14,
    "output": 0.57,
    "quantization": "fp8",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.2475,
    "variants": 1
   },
   "spreadRatio": 1.9,
   "sameQuantisationEverywhere": true,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.08,
     "output": 0.28,
     "quantization": "fp8",
     "contextLength": 40960,
     "uptime30m": 100,
     "blended": 0.13,
     "variants": 1
    },
    {
     "host": "Tenstorrent",
     "input": 0.14,
     "output": 0.4,
     "quantization": "fp8",
     "contextLength": 20000,
     "uptime30m": 100,
     "blended": 0.205,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.14,
     "output": 0.57,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.2475,
     "variants": 1
    }
   ]
  },
  {
   "model": "meta-llama/llama-4-maverick",
   "huggingFaceId": "meta-llama/Llama-4-Maverick-17B-128E-Instruct",
   "hosts": 3,
   "cheapest": {
    "host": "DigitalOcean",
    "input": 0.1875,
    "output": 0.6525,
    "quantization": "unknown",
    "contextLength": 128000,
    "uptime30m": 100,
    "blended": 0.30375,
    "variants": 1
   },
   "dearest": {
    "host": "Google",
    "input": 0.35,
    "output": 1.15,
    "quantization": "unknown",
    "contextLength": 524288,
    "uptime30m": null,
    "blended": 0.55,
    "variants": 1
   },
   "spreadRatio": 1.81,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DigitalOcean",
     "input": 0.1875,
     "output": 0.6525,
     "quantization": "unknown",
     "contextLength": 128000,
     "uptime30m": 100,
     "blended": 0.30375,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.35,
     "output": 1,
     "quantization": "fp8",
     "contextLength": 524288,
     "uptime30m": 100,
     "blended": 0.5125,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.35,
     "output": 1.15,
     "quantization": "unknown",
     "contextLength": 524288,
     "uptime30m": null,
     "blended": 0.55,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3-vl-30b-a3b-instruct",
   "huggingFaceId": "Qwen/Qwen3-VL-30B-A3B-Instruct",
   "hosts": 3,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.15,
    "output": 0.6,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.2625,
    "variants": 1
   },
   "dearest": {
    "host": "SiliconFlow",
    "input": 0.29,
    "output": 1,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 33.33333333333333,
    "blended": 0.4675,
    "variants": 1
   },
   "spreadRatio": 1.78,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.15,
     "output": 0.6,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.2625,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.2,
     "output": 0.7,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 93.33333333333333,
     "blended": 0.325,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.29,
     "output": 1,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 33.33333333333333,
     "blended": 0.4675,
     "variants": 1
    }
   ]
  },
  {
   "model": "google/gemma-3-27b-it",
   "huggingFaceId": "google/gemma-3-27b-it",
   "hosts": 5,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.08,
    "output": 0.16,
    "quantization": "fp8",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.1,
    "variants": 1
   },
   "dearest": {
    "host": "Parasail",
    "input": 0.08,
    "output": 0.45,
    "quantization": "fp8",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.1725,
    "variants": 1
   },
   "spreadRatio": 1.73,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.08,
     "output": 0.16,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.1,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.119,
     "output": 0.2,
     "quantization": "bf16",
     "contextLength": 98304,
     "uptime30m": 100,
     "blended": 0.13925,
     "variants": 1
    },
    {
     "host": "NextBit",
     "input": 0.08,
     "output": 0.35,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.1475,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 0.1,
     "output": 0.3,
     "quantization": "fp8",
     "contextLength": 110000,
     "uptime30m": 100,
     "blended": 0.15,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.08,
     "output": 0.45,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.1725,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3-vl-235b-a22b-instruct",
   "huggingFaceId": "Qwen/Qwen3-VL-235B-A22B-Instruct",
   "hosts": 3,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.2,
    "output": 0.88,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.37,
    "variants": 1
   },
   "dearest": {
    "host": "Parasail",
    "input": 0.21,
    "output": 1.9,
    "quantization": "fp8",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.6325,
    "variants": 1
   },
   "spreadRatio": 1.71,
   "sameQuantisationEverywhere": true,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.2,
     "output": 0.88,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.37,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.21,
     "output": 1.9,
     "quantization": "fp8",
     "contextLength": 128000,
     "uptime30m": 100,
     "blended": 0.6325,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.21,
     "output": 1.9,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.6325,
     "variants": 1
    }
   ]
  },
  {
   "model": "nvidia/nemotron-3-ultra-550b-a55b",
   "huggingFaceId": "nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B-BF16",
   "hosts": 3,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.4,
    "output": 1.76,
    "quantization": "fp4",
    "contextLength": 262144,
    "uptime30m": null,
    "blended": 0.74,
    "variants": 2
   },
   "dearest": {
    "host": "Venice",
    "input": 0.625,
    "output": 3.125,
    "quantization": "fp8",
    "contextLength": 256000,
    "uptime30m": 95.23809523809523,
    "blended": 1.25,
    "variants": 1
   },
   "spreadRatio": 1.69,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.4,
     "output": 1.76,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 0.74,
     "variants": 2
    },
    {
     "host": "BaseTen",
     "input": 0.6,
     "output": 2.4,
     "quantization": "fp4",
     "contextLength": 202800,
     "uptime30m": 100,
     "blended": 1.05,
     "variants": 2
    },
    {
     "host": "Venice",
     "input": 0.625,
     "output": 3.125,
     "quantization": "fp8",
     "contextLength": 256000,
     "uptime30m": 95.23809523809523,
     "blended": 1.25,
     "variants": 1
    }
   ]
  },
  {
   "model": "z-ai/glm-5",
   "huggingFaceId": "zai-org/GLM-5",
   "hosts": 8,
   "cheapest": {
    "host": "StreamLake",
    "input": 0.6,
    "output": 1.92,
    "quantization": "fp8",
    "contextLength": 198000,
    "uptime30m": 100,
    "blended": 0.93,
    "variants": 1
   },
   "dearest": {
    "host": "Z.AI",
    "input": 1,
    "output": 3.2,
    "quantization": "fp8",
    "contextLength": 202752,
    "uptime30m": 100,
    "blended": 1.55,
    "variants": 1
   },
   "spreadRatio": 1.67,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "StreamLake",
     "input": 0.6,
     "output": 1.92,
     "quantization": "fp8",
     "contextLength": 198000,
     "uptime30m": 100,
     "blended": 0.93,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.6,
     "output": 1.92,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 0.93,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 0.7,
     "output": 2.24,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 1.085,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.95,
     "output": 2.55,
     "quantization": "fp8",
     "contextLength": 204800,
     "uptime30m": 100,
     "blended": 1.35,
     "variants": 1
    },
    {
     "host": "Amazon Bedrock",
     "input": 1,
     "output": 3.2,
     "quantization": "unknown",
     "contextLength": 202752,
     "uptime30m": null,
     "blended": 1.55,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 1,
     "output": 3.2,
     "quantization": "fp8",
     "contextLength": 198000,
     "uptime30m": 100,
     "blended": 1.55,
     "variants": 2
    },
    {
     "host": "Novita",
     "input": 1,
     "output": 3.2,
     "quantization": "fp8",
     "contextLength": 202800,
     "uptime30m": 100,
     "blended": 1.55,
     "variants": 1
    },
    {
     "host": "Z.AI",
     "input": 1,
     "output": 3.2,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 1.55,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3.6-27b",
   "huggingFaceId": "Qwen/Qwen3.6-27B",
   "hosts": 7,
   "cheapest": {
    "host": "Reka",
    "input": 0.2,
    "output": 2,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.65,
    "variants": 1
   },
   "dearest": {
    "host": "Venice",
    "input": 0.325,
    "output": 3.25,
    "quantization": "fp8",
    "contextLength": 256000,
    "uptime30m": 100,
    "blended": 1.05625,
    "variants": 1
   },
   "spreadRatio": 1.62,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Reka",
     "input": 0.2,
     "output": 2,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.65,
     "variants": 1
    },
    {
     "host": "Chutes",
     "input": 0.25,
     "output": 2,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.6875,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.45,
     "output": 2.7,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.0125,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.3,
     "output": 3.2,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.025,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.32,
     "output": 3.2,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.04,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.32,
     "output": 3.25,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 96.66666666666667,
     "blended": 1.0525,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.325,
     "output": 3.25,
     "quantization": "fp8",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 1.05625,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3.5-27b",
   "huggingFaceId": "Qwen/Qwen3.5-27B",
   "hosts": 6,
   "cheapest": {
    "host": "Alibaba",
    "input": 0.195,
    "output": 1.56,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.53625,
    "variants": 1
   },
   "dearest": {
    "host": "DeepInfra",
    "input": 0.26,
    "output": 2.6,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.845,
    "variants": 1
   },
   "spreadRatio": 1.58,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Alibaba",
     "input": 0.195,
     "output": 1.56,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.53625,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.25,
     "output": 2,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 0.6875,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.27,
     "output": 2.16,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.7425,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.3,
     "output": 2.4,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 75,
     "blended": 0.825,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.3,
     "output": 2.4,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.825,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.26,
     "output": 2.6,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.845,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3.5-122b-a10b",
   "huggingFaceId": "Qwen/Qwen3.5-122B-A10B",
   "hosts": 4,
   "cheapest": {
    "host": "SiliconFlow",
    "input": 0.26,
    "output": 2.08,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.715,
    "variants": 1
   },
   "dearest": {
    "host": "Novita",
    "input": 0.4,
    "output": 3.2,
    "quantization": "bf16",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 1.1,
    "variants": 1
   },
   "spreadRatio": 1.54,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "SiliconFlow",
     "input": 0.26,
     "output": 2.08,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.715,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.26,
     "output": 2.08,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.715,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.3,
     "output": 2.4,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.825,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.4,
     "output": 3.2,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.1,
     "variants": 1
    }
   ]
  },
  {
   "model": "tencent/hy3",
   "huggingFaceId": "tencent/Hy3",
   "hosts": 5,
   "cheapest": {
    "host": "Tencent",
    "input": 0.132,
    "output": 0.528,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.231,
    "variants": 1
   },
   "dearest": {
    "host": "AtlasCloud",
    "input": 0.2,
    "output": 0.8,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.35,
    "variants": 1
   },
   "spreadRatio": 1.52,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Tencent",
     "input": 0.132,
     "output": 0.528,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.231,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.14,
     "output": 0.58,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.25,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.14,
     "output": 0.58,
     "quantization": "bf16",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.25,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 0.15,
     "output": 0.64,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.2725,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.2,
     "output": 0.8,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.35,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3-next-80b-a3b-instruct",
   "huggingFaceId": "Qwen/Qwen3-Next-80B-A3B-Instruct",
   "hosts": 3,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.072,
    "output": 0.88,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": null,
    "blended": 0.274,
    "variants": 3
   },
   "dearest": {
    "host": "Google",
    "input": 0.15,
    "output": 1.2,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 0,
    "blended": 0.4125,
    "variants": 1
   },
   "spreadRatio": 1.51,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.072,
     "output": 0.88,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 0.274,
     "variants": 3
    },
    {
     "host": "Parasail",
     "input": 0.1,
     "output": 1.1,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.35,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.15,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 0,
     "blended": 0.4125,
     "variants": 1
    }
   ]
  },
  {
   "model": "z-ai/glm-5.1",
   "huggingFaceId": "zai-org/GLM-5.1",
   "hosts": 13,
   "cheapest": {
    "host": "StreamLake",
    "input": 0.966,
    "output": 3.036,
    "quantization": "fp8",
    "contextLength": 200000,
    "uptime30m": 100,
    "blended": 1.4835,
    "variants": 1
   },
   "dearest": {
    "host": "Venice",
    "input": 1.4014,
    "output": 4.4044,
    "quantization": "fp8",
    "contextLength": 200000,
    "uptime30m": null,
    "blended": 2.15215,
    "variants": 1
   },
   "spreadRatio": 1.45,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "StreamLake",
     "input": 0.966,
     "output": 3.036,
     "quantization": "fp8",
     "contextLength": 200000,
     "uptime30m": 100,
     "blended": 1.4835,
     "variants": 1
    },
    {
     "host": "Chutes",
     "input": 0.98,
     "output": 3.08,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 1.505,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 1.19,
     "output": 3.74,
     "quantization": "fp8",
     "contextLength": 204800,
     "uptime30m": 100,
     "blended": 1.8275,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 1.26,
     "output": 3.96,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": null,
     "blended": 1.935,
     "variants": 1
    },
    {
     "host": "Phala",
     "input": 1.21,
     "output": 4.2,
     "quantization": "unknown",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 1.9575,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 1.33,
     "output": 4.18,
     "quantization": "fp8",
     "contextLength": 202745,
     "uptime30m": null,
     "blended": 2.0425,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 1.38,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 204800,
     "uptime30m": null,
     "blended": 2.135,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Baidu",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": null,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": null,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Friendli",
     "input": 1.4,
     "output": 4.4,
     "quantization": "unknown",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Z.AI",
     "input": 1.4,
     "output": 4.4,
     "quantization": "fp8",
     "contextLength": 202752,
     "uptime30m": 95.83333333333334,
     "blended": 2.15,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 1.4014,
     "output": 4.4044,
     "quantization": "fp8",
     "contextLength": 200000,
     "uptime30m": null,
     "blended": 2.15215,
     "variants": 1
    }
   ]
  },
  {
   "model": "z-ai/glm-4.6",
   "huggingFaceId": "zai-org/GLM-4.6",
   "hosts": 4,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.4,
    "output": 1.6,
    "quantization": "fp4",
    "contextLength": 202752,
    "uptime30m": null,
    "blended": 0.7,
    "variants": 2
   },
   "dearest": {
    "host": "Z.AI",
    "input": 0.6,
    "output": 2.2,
    "quantization": "fp4",
    "contextLength": 202752,
    "uptime30m": 100,
    "blended": 1,
    "variants": 1
   },
   "spreadRatio": 1.43,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.4,
     "output": 1.6,
     "quantization": "fp4",
     "contextLength": 202752,
     "uptime30m": null,
     "blended": 0.7,
     "variants": 2
    },
    {
     "host": "Venice",
     "input": 0.43,
     "output": 1.75,
     "quantization": "fp4",
     "contextLength": 198000,
     "uptime30m": 100,
     "blended": 0.76,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.55,
     "output": 2.2,
     "quantization": "bf16",
     "contextLength": 204800,
     "uptime30m": 100,
     "blended": 0.9625,
     "variants": 1
    },
    {
     "host": "Z.AI",
     "input": 0.6,
     "output": 2.2,
     "quantization": "fp4",
     "contextLength": 202752,
     "uptime30m": 100,
     "blended": 1,
     "variants": 1
    }
   ]
  },
  {
   "model": "reflection/beam-1.0-501b-a23b",
   "huggingFaceId": "reflection/Beam-1.0-501B-A23B",
   "hosts": 6,
   "cheapest": {
    "host": "Io Net",
    "input": 0.24,
    "output": 0.98,
    "quantization": "fp4",
    "contextLength": 1048320,
    "uptime30m": 100,
    "blended": 0.425,
    "variants": 1
   },
   "dearest": {
    "host": "Together",
    "input": 0.4,
    "output": 1.2,
    "quantization": "unknown",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.6,
    "variants": 1
   },
   "spreadRatio": 1.41,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Io Net",
     "input": 0.24,
     "output": 0.98,
     "quantization": "fp4",
     "contextLength": 1048320,
     "uptime30m": 100,
     "blended": 0.425,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 0.25,
     "output": 1,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 0.4375,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.3,
     "output": 1,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.475,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.4,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 0.6,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 0.4,
     "output": 1.2,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 0.6,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 0.4,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.6,
     "variants": 1
    }
   ]
  },
  {
   "model": "mistralai/mistral-small-3.2-24b-instruct",
   "huggingFaceId": "mistralai/Mistral-Small-3.2-24B-Instruct-2506",
   "hosts": 4,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.075,
    "output": 0.2,
    "quantization": "fp8",
    "contextLength": 128000,
    "uptime30m": 100,
    "blended": 0.10625,
    "variants": 1
   },
   "dearest": {
    "host": "Mistral",
    "input": 0.1,
    "output": 0.3,
    "quantization": "unknown",
    "contextLength": 32768,
    "uptime30m": 100,
    "blended": 0.15,
    "variants": 1
   },
   "spreadRatio": 1.41,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.075,
     "output": 0.2,
     "quantization": "fp8",
     "contextLength": 128000,
     "uptime30m": 100,
     "blended": 0.10625,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.09375,
     "output": 0.25,
     "quantization": "fp8",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 0.132813,
     "variants": 1
    },
    {
     "host": "Parasail",
     "input": 0.09,
     "output": 0.3,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.1425,
     "variants": 1
    },
    {
     "host": "Mistral",
     "input": 0.1,
     "output": 0.3,
     "quantization": "unknown",
     "contextLength": 32768,
     "uptime30m": 100,
     "blended": 0.15,
     "variants": 1
    }
   ]
  },
  {
   "model": "z-ai/glm-4.5-air",
   "huggingFaceId": "zai-org/GLM-4.5-Air",
   "hosts": 3,
   "cheapest": {
    "host": "Novita",
    "input": 0.13,
    "output": 0.85,
    "quantization": "bf16",
    "contextLength": 131072,
    "uptime30m": 93.33333333333333,
    "blended": 0.31,
    "variants": 1
   },
   "dearest": {
    "host": "Z.AI",
    "input": 0.2,
    "output": 1.1,
    "quantization": "fp8",
    "contextLength": 131072,
    "uptime30m": 100,
    "blended": 0.425,
    "variants": 1
   },
   "spreadRatio": 1.37,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Novita",
     "input": 0.13,
     "output": 0.85,
     "quantization": "bf16",
     "contextLength": 131072,
     "uptime30m": 93.33333333333333,
     "blended": 0.31,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.14,
     "output": 0.86,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.32,
     "variants": 1
    },
    {
     "host": "Z.AI",
     "input": 0.2,
     "output": 1.1,
     "quantization": "fp8",
     "contextLength": 131072,
     "uptime30m": 100,
     "blended": 0.425,
     "variants": 1
    }
   ]
  },
  {
   "model": "xiaomi/mimo-v2.6-pro",
   "huggingFaceId": "XiaomiMiMo/MiMo-V2.6-Pro-RL",
   "hosts": 5,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.387,
    "output": 0.783,
    "quantization": "fp8",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 0.486,
    "variants": 1
   },
   "dearest": {
    "host": "StreamLake",
    "input": 0.522,
    "output": 1.044,
    "quantization": "unknown",
    "contextLength": 1000000,
    "uptime30m": null,
    "blended": 0.6525,
    "variants": 1
   },
   "spreadRatio": 1.34,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.387,
     "output": 0.783,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.486,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.435,
     "output": 0.87,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.54375,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.435,
     "output": 0.87,
     "quantization": "bf16",
     "contextLength": 1050000,
     "uptime30m": 100,
     "blended": 0.54375,
     "variants": 1
    },
    {
     "host": "Xiaomi",
     "input": 0.435,
     "output": 0.87,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 0.54375,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.522,
     "output": 1.044,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": null,
     "blended": 0.6525,
     "variants": 1
    }
   ]
  },
  {
   "model": "moonshotai/kimi-k2.7-code",
   "huggingFaceId": "moonshotai/Kimi-K2.7-Code",
   "hosts": 11,
   "cheapest": {
    "host": "Inceptron",
    "input": 0.6725,
    "output": 3.1,
    "quantization": "int4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 1.279375,
    "variants": 1
   },
   "dearest": {
    "host": "Moonshot AI",
    "input": 0.95,
    "output": 4,
    "quantization": "int4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 1.7125,
    "variants": 2
   },
   "spreadRatio": 1.34,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Inceptron",
     "input": 0.6725,
     "output": 3.1,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.279375,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.7125,
     "output": 3,
     "quantization": "unknown",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 1.284375,
     "variants": 1
    },
    {
     "host": "CoreWeave",
     "input": 0.71,
     "output": 3.5,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 1.4075,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.75,
     "output": 3.5,
     "quantization": "int4",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 1.4375,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.85916,
     "output": 3.8,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.59437,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.912,
     "output": 3.84,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 1.644,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 0.95,
     "output": 4,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.7125,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.95,
     "output": 4,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.7125,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 0.95,
     "output": 4,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 1.7125,
     "variants": 1
    },
    {
     "host": "Cloudflare",
     "input": 0.95,
     "output": 4,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.7125,
     "variants": 1
    },
    {
     "host": "Moonshot AI",
     "input": 0.95,
     "output": 4,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.7125,
     "variants": 2
    }
   ]
  },
  {
   "model": "qwen/qwen3-coder",
   "huggingFaceId": "Qwen/Qwen3-Coder-480B-A35B-Instruct",
   "hosts": 3,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.3,
    "output": 1,
    "quantization": "fp4",
    "contextLength": 262144,
    "uptime30m": 35.714285714285715,
    "blended": 0.475,
    "variants": 1
   },
   "dearest": {
    "host": "Venice",
    "input": 0.35,
    "output": 1.5,
    "quantization": "fp8",
    "contextLength": 256000,
    "uptime30m": 35.714285714285715,
    "blended": 0.6375,
    "variants": 1
   },
   "spreadRatio": 1.34,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.3,
     "output": 1,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": 35.714285714285715,
     "blended": 0.475,
     "variants": 1
    },
    {
     "host": "Google",
     "input": 0.22,
     "output": 1.8,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 0,
     "blended": 0.615,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.35,
     "output": 1.5,
     "quantization": "fp8",
     "contextLength": 256000,
     "uptime30m": 35.714285714285715,
     "blended": 0.6375,
     "variants": 1
    }
   ]
  },
  {
   "model": "moonshotai/kimi-k2.5",
   "huggingFaceId": "moonshotai/Kimi-K2.5",
   "hosts": 5,
   "cheapest": {
    "host": "AtlasCloud",
    "input": 0.49,
    "output": 2.5,
    "quantization": "int4",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.9925,
    "variants": 1
   },
   "dearest": {
    "host": "Venice",
    "input": 0.532,
    "output": 3.325,
    "quantization": "unknown",
    "contextLength": 256000,
    "uptime30m": 100,
    "blended": 1.23025,
    "variants": 1
   },
   "spreadRatio": 1.24,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "AtlasCloud",
     "input": 0.49,
     "output": 2.5,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.9925,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.5,
     "output": 2.5,
     "quantization": "int4",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.57,
     "output": 2.85,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.14,
     "variants": 1
    },
    {
     "host": "Amazon Bedrock",
     "input": 0.6,
     "output": 3,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.2,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 0.532,
     "output": 3.325,
     "quantization": "unknown",
     "contextLength": 256000,
     "uptime30m": 100,
     "blended": 1.23025,
     "variants": 1
    }
   ]
  },
  {
   "model": "minimax/minimax-m2.5",
   "huggingFaceId": "MiniMaxAI/MiniMax-M2.5",
   "hosts": 7,
   "cheapest": {
    "host": "Venice",
    "input": 0.27,
    "output": 0.95,
    "quantization": "unknown",
    "contextLength": 198000,
    "uptime30m": 80,
    "blended": 0.44,
    "variants": 1
   },
   "dearest": {
    "host": "Minimax",
    "input": 0.3,
    "output": 1.2,
    "quantization": "fp8",
    "contextLength": 204800,
    "uptime30m": null,
    "blended": 0.525,
    "variants": 2
   },
   "spreadRatio": 1.19,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Venice",
     "input": 0.27,
     "output": 0.95,
     "quantization": "unknown",
     "contextLength": 198000,
     "uptime30m": 80,
     "blended": 0.44,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.27,
     "output": 1.08,
     "quantization": "unknown",
     "contextLength": 200000,
     "uptime30m": 100,
     "blended": 0.4725,
     "variants": 1
    },
    {
     "host": "AtlasCloud",
     "input": 0.295,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 196608,
     "uptime30m": 100,
     "blended": 0.52125,
     "variants": 1
    },
    {
     "host": "GMICloud",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 196608,
     "uptime30m": null,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "DigitalOcean",
     "input": 0.3,
     "output": 1.2,
     "quantization": "unknown",
     "contextLength": 65536,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 204800,
     "uptime30m": 100,
     "blended": 0.525,
     "variants": 1
    },
    {
     "host": "Minimax",
     "input": 0.3,
     "output": 1.2,
     "quantization": "fp8",
     "contextLength": 204800,
     "uptime30m": null,
     "blended": 0.525,
     "variants": 2
    }
   ]
  },
  {
   "model": "deepseek/deepseek-r1-0528",
   "huggingFaceId": "deepseek-ai/DeepSeek-R1-0528",
   "hosts": 3,
   "cheapest": {
    "host": "DeepInfra",
    "input": 0.5,
    "output": 2.15,
    "quantization": "fp4",
    "contextLength": 163840,
    "uptime30m": 100,
    "blended": 0.9125,
    "variants": 1
   },
   "dearest": {
    "host": "StreamLake",
    "input": 0.571,
    "output": 2.286,
    "quantization": "unknown",
    "contextLength": 128000,
    "uptime30m": 100,
    "blended": 0.99975,
    "variants": 1
   },
   "spreadRatio": 1.1,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DeepInfra",
     "input": 0.5,
     "output": 2.15,
     "quantization": "fp4",
     "contextLength": 163840,
     "uptime30m": 100,
     "blended": 0.9125,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.5,
     "output": 2.18,
     "quantization": "fp8",
     "contextLength": 163840,
     "uptime30m": 100,
     "blended": 0.92,
     "variants": 1
    },
    {
     "host": "StreamLake",
     "input": 0.571,
     "output": 2.286,
     "quantization": "unknown",
     "contextLength": 128000,
     "uptime30m": 100,
     "blended": 0.99975,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3-30b-a3b-instruct-2507",
   "huggingFaceId": "Qwen/Qwen3-30B-A3B-Instruct-2507",
   "hosts": 3,
   "cheapest": {
    "host": "DekaLLM",
    "input": 0.09,
    "output": 0.3,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.1425,
    "variants": 1
   },
   "dearest": {
    "host": "Nebius",
    "input": 0.1,
    "output": 0.3,
    "quantization": "fp8",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 0.15,
    "variants": 1
   },
   "spreadRatio": 1.05,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "DekaLLM",
     "input": 0.09,
     "output": 0.3,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1425,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.09,
     "output": 0.3,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.1425,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 0.1,
     "output": 0.3,
     "quantization": "fp8",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 0.15,
     "variants": 1
    }
   ]
  },
  {
   "model": "meta/muse-spark-open",
   "huggingFaceId": "meta-models/Muse-Spark-Open-816B-A42B",
   "hosts": 6,
   "cheapest": {
    "host": "Makora",
    "input": 1.25,
    "output": 4.25,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 2,
    "variants": 1
   },
   "dearest": {
    "host": "BaseTen",
    "input": 1.25,
    "output": 4.25,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": null,
    "blended": 2,
    "variants": 1
   },
   "spreadRatio": 1,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Makora",
     "input": 1.25,
     "output": 4.25,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2,
     "variants": 1
    },
    {
     "host": "Modal",
     "input": 1.25,
     "output": 4.25,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 2,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 1.25,
     "output": 4.25,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 2,
     "variants": 1
    },
    {
     "host": "Nebius",
     "input": 1.25,
     "output": 4.25,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 2,
     "variants": 1
    },
    {
     "host": "Fireworks",
     "input": 1.25,
     "output": 4.25,
     "quantization": "unknown",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 2,
     "variants": 1
    },
    {
     "host": "BaseTen",
     "input": 1.25,
     "output": 4.25,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 2,
     "variants": 1
    }
   ]
  },
  {
   "model": "stepfun/step-5-preview",
   "huggingFaceId": "stepfun-ai/Step-5-Preview",
   "hosts": 4,
   "cheapest": {
    "host": "InferenceNet",
    "input": 1,
    "output": 2.7,
    "quantization": "fp4",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 1.425,
    "variants": 2
   },
   "dearest": {
    "host": "StepFun",
    "input": 1,
    "output": 2.7,
    "quantization": "fp8",
    "contextLength": 1000000,
    "uptime30m": 100,
    "blended": 1.425,
    "variants": 1
   },
   "spreadRatio": 1,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "InferenceNet",
     "input": 1,
     "output": 2.7,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.425,
     "variants": 2
    },
    {
     "host": "Featherless",
     "input": 1,
     "output": 2.7,
     "quantization": "fp4",
     "contextLength": 262144,
     "uptime30m": null,
     "blended": 1.425,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 1,
     "output": 2.7,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": null,
     "blended": 1.425,
     "variants": 1
    },
    {
     "host": "StepFun",
     "input": 1,
     "output": 2.7,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 1.425,
     "variants": 1
    }
   ]
  },
  {
   "model": "tencent/hy4-preview",
   "huggingFaceId": "tencent/Hy4-preview",
   "hosts": 5,
   "cheapest": {
    "host": "Cohere",
    "input": 0.834,
    "output": 2.501,
    "quantization": "unknown",
    "contextLength": 262144,
    "uptime30m": 100,
    "blended": 1.25075,
    "variants": 1
   },
   "dearest": {
    "host": "Tencent",
    "input": 0.834,
    "output": 2.501,
    "quantization": "fp8",
    "contextLength": 1048576,
    "uptime30m": 100,
    "blended": 1.25075,
    "variants": 1
   },
   "spreadRatio": 1,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Cohere",
     "input": 0.834,
     "output": 2.501,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 1.25075,
     "variants": 1
    },
    {
     "host": "DeepInfra",
     "input": 0.834,
     "output": 2.501,
     "quantization": "fp4",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.25075,
     "variants": 1
    },
    {
     "host": "Novita",
     "input": 0.834,
     "output": 2.501,
     "quantization": "fp8",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 1.25075,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 0.834,
     "output": 2.501,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.25075,
     "variants": 1
    },
    {
     "host": "Tencent",
     "input": 0.834,
     "output": 2.501,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 1.25075,
     "variants": 1
    }
   ]
  },
  {
   "model": "qwen/qwen3.8-2.4t-a95b",
   "huggingFaceId": "Qwen/Qwen3.8-2.4T-A95B",
   "hosts": 5,
   "cheapest": {
    "host": "Novita",
    "input": 2,
    "output": 6,
    "quantization": "unknown",
    "contextLength": 1000000,
    "uptime30m": 100,
    "blended": 3,
    "variants": 1
   },
   "dearest": {
    "host": "Together",
    "input": 2,
    "output": 6,
    "quantization": "unknown",
    "contextLength": 1010000,
    "uptime30m": 100,
    "blended": 3,
    "variants": 1
   },
   "spreadRatio": 1,
   "sameQuantisationEverywhere": false,
   "endpoints": [
    {
     "host": "Novita",
     "input": 2,
     "output": 6,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 3,
     "variants": 1
    },
    {
     "host": "Alibaba",
     "input": 2,
     "output": 6,
     "quantization": "unknown",
     "contextLength": 1000000,
     "uptime30m": 100,
     "blended": 3,
     "variants": 1
    },
    {
     "host": "SiliconFlow",
     "input": 2,
     "output": 6,
     "quantization": "fp8",
     "contextLength": 1048576,
     "uptime30m": 100,
     "blended": 3,
     "variants": 1
    },
    {
     "host": "Venice",
     "input": 2,
     "output": 6,
     "quantization": "unknown",
     "contextLength": 262144,
     "uptime30m": 100,
     "blended": 3,
     "variants": 1
    },
    {
     "host": "Together",
     "input": 2,
     "output": 6,
     "quantization": "unknown",
     "contextLength": 1010000,
     "uptime30m": 100,
     "blended": 3,
     "variants": 1
    }
   ]
  }
 ]
}
