{
  "_provenance": {
    "rev": "litellm@2026-07-14#179affb",
    "pricesAsOf": "2026-07-14",
    "url": "https://tokentriage.com/models/lambda-ai-lambda-ai-llama-4-maverick-17b-128e-instruct-fp8",
    "sources": {
      "pricing": "TokenTriage resolved price artifact (LiteLLM-derived) — the rates the product bills from",
      "capabilities": "LiteLLM model_prices_and_context_window.json (MIT © Berri AI)",
      "metadata": "models.dev (MIT)",
      "benchmarks": "Epoch AI — \"AI Benchmarking Hub\" (CC-BY (Epoch AI)); some rows carry an upstream leaderboard license (per benchmark.sourceLicense)",
      "elo": "LMArena — human-preference Elo (CC-BY-4.0); LMArena Leaderboard (lmarena.ai) — lmarena-ai/leaderboard-dataset on Hugging Face, CC BY 4.0.",
      "capabilitiesIndex": "Epoch AI — Capabilities Index (CC-BY (Epoch AI))",
      "modelFacts": "Epoch AI — Notable AI Models (CC-BY (Epoch AI))",
      "openWeightsSpec": "Hugging Face Hub — per-repo model license (SPDX where stated) · facts via HF API",
      "crosscheck": "models.dev v2 (MIT) · Portkey (MIT) · TrueFoundry (MIT) — capability fill + first-party price corroboration/conflict"
    },
    "honesty": "unknown is null, never 0. Free to reuse with attribution. Elo is human preference, not correctness; ECI is Epoch’s relative re-fit scale; both carry 95% CIs."
  },
  "id": "lambda-ai-lambda-ai-llama-4-maverick-17b-128e-instruct-fp8",
  "key": "lambda_ai/lambda_ai/llama-4-maverick-17b-128e-instruct-fp8",
  "name": "lambda_ai/llama-4-maverick-17b-128e-instruct-fp8",
  "provider": "lambda_ai",
  "mode": "chat",
  "pricing": {
    "input": 5e-8,
    "output": 1e-7,
    "cacheRead": null,
    "cacheWrite": null,
    "cacheWrite5m": null,
    "cacheWrite1h": null,
    "reasoning": null,
    "batchInput": null,
    "batchOutput": null,
    "perImage": null,
    "perSecond": null,
    "perChar": null,
    "tiers": []
  },
  "context": {
    "maxInput": 131072,
    "maxOutput": 8192
  },
  "costPerRequest_1kIn_500out_usd": 0.0001,
  "capabilities": {
    "functionCalling": "yes",
    "parallelToolCalls": "yes",
    "vision": "unknown",
    "structuredOutput": "unknown",
    "reasoning": "unknown",
    "promptCaching": "unknown",
    "streaming": "unknown",
    "webSearch": "unknown",
    "pdfInput": "unknown",
    "audioInput": "unknown",
    "audioOutput": "unknown",
    "computerUse": "unknown"
  },
  "deprecated": null,
  "knowledgeCutoff": "2024-08",
  "releaseDate": "2025-04-05",
  "openWeights": true,
  "benchmarks": [
    {
      "id": "gpqa",
      "label": "GPQA Diamond",
      "desc": "graduate-level science reasoning",
      "domain": "reasoning",
      "score": 0.5597643097643098,
      "source": "Epoch evaluations",
      "sourceLicense": "CC-BY (Epoch AI)",
      "optimized": true,
      "providerReported": false,
      "epochModel": "Llama 4 Maverick",
      "date": "2025-04-05"
    },
    {
      "id": "math",
      "label": "MATH Lvl 5",
      "desc": "hard competition math",
      "domain": "math",
      "score": 0.7301737160120846,
      "source": "Epoch evaluations",
      "sourceLicense": "CC-BY (Epoch AI)",
      "optimized": true,
      "providerReported": false,
      "epochModel": "Llama 4 Maverick",
      "date": "2025-04-05"
    },
    {
      "id": "aime",
      "label": "AIME (OTIS mock)",
      "desc": "olympiad math",
      "domain": "math",
      "score": 0.20476031587142693,
      "source": "Epoch evaluations",
      "sourceLicense": "CC-BY (Epoch AI)",
      "optimized": true,
      "providerReported": false,
      "epochModel": "Llama 4 Maverick",
      "date": "2025-04-05"
    },
    {
      "id": "frontiermath",
      "label": "FrontierMath",
      "desc": "research-level math",
      "domain": "math",
      "score": 0.012099213551119124,
      "source": "Epoch evaluations",
      "sourceLicense": "CC-BY (Epoch AI)",
      "optimized": true,
      "providerReported": false,
      "epochModel": "Llama 4 Maverick",
      "date": "2025-04-05"
    }
  ],
  "elo": null,
  "capabilitiesIndex": null,
  "modelFacts": null,
  "openWeightsSpec": {
    "repo": "meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
    "paramsTotal": 401649841664,
    "paramsByDtype": {
      "BF16": 15102785024,
      "F8_E4M3": 386547056640
    },
    "activeParams": null,
    "architecture": "Llama4ForConditionalGeneration",
    "modelType": "llama4",
    "numExperts": null,
    "expertsPerTok": null,
    "nativeDtype": "F8_E4M3",
    "footprintBytes": 416752626688,
    "preQuantized": true,
    "licenseId": "other",
    "commercialUse": "conditional",
    "asOf": "2026-07-30",
    "sourceUrl": "https://huggingface.co/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8",
    "sha": "94125d2bd83076b21eed33119525e29eaf3894f4",
    "lastModified": "2025-05-22T23:46:03.000Z",
    "quantMethod": "compressed-tensors",
    "licenseName": "llama4"
  },
  "capabilitiesCrosscheck": {
    "functionCalling": {
      "resolved": "yes",
      "corroborated": true,
      "sources": [
        {
          "name": "models.dev",
          "value": "yes",
          "asOf": "2026-07-29"
        }
      ]
    },
    "vision": {
      "resolved": "yes",
      "sources": [
        {
          "name": "models.dev",
          "value": "yes",
          "asOf": "2026-07-29"
        }
      ],
      "derived": true
    },
    "reasoning": {
      "resolved": "no",
      "sources": [
        {
          "name": "models.dev",
          "value": "no",
          "asOf": "2026-07-29"
        }
      ],
      "semantic": "models.dev marks a reasoning MODE; LiteLLM tracks billed reasoning tokens — these can differ."
    }
  },
  "priceCrosscheck": null,
  "embeddingQuality": null
}