{
  "id": "provider:replicate",
  "_kind": "Provider",
  "_file": "compute/providers/replicate.yaml",
  "_cluster": "compute",
  "attributes": {
    "versionRange": ">=2024-01-01",
    "displayName": "Replicate",
    "vendor": "Replicate",
    "authMethods": [
      "api-key"
    ],
    "authMethodNotes": "`Authorization: Token <api-key>` header (Replicate-native), with an\nOpenAI-compatible chat endpoint also available for many models.\n",
    "endpoints": {
      "base": "https://api.replicate.com/v1",
      "predictions": "https://api.replicate.com/v1/predictions",
      "models": "https://api.replicate.com/v1/models"
    },
    "pricing": "Pay-per-token or pay-per-second of GPU time depending on the model;\nsee https://replicate.com/pricing\n",
    "pricingTiers": [
      {
        "name": "serverless",
        "rateLimit": "Per-account RPS caps; queue when saturated",
        "priceMultiplier": 1,
        "description": "Pay-per-call serverless inference."
      },
      {
        "name": "dedicated",
        "rateLimit": "Provisioned hardware; throughput per deployment",
        "priceMultiplier": 1,
        "description": "Dedicated deployments — reserved GPU capacity."
      }
    ],
    "rateLimitSignalingProtocol": "HTTP 429 with `retry-after` for pay-as-you-go; queued predictions return\na status URL that callers poll.\n",
    "dataResidencyOptions": [
      "us"
    ],
    "vendorFeatures": [
      "capability:streaming"
    ],
    "slaTier": "replicate-no-public-sla",
    "regions": [
      "global"
    ]
  },
  "outgoingEdges": [
    {
      "from": "provider:replicate",
      "to": "layer:2-provider",
      "kind": "realizes",
      "attributes": {}
    },
    {
      "from": "provider:replicate",
      "to": "model:llama-3-1-405b-instruct@current",
      "kind": "serves"
    },
    {
      "from": "provider:replicate",
      "to": "model:llama-3-1-70b-instruct@current",
      "kind": "serves"
    },
    {
      "from": "provider:replicate",
      "to": "model:llama-3-3-70b-instruct@current",
      "kind": "serves"
    },
    {
      "from": "provider:replicate",
      "to": "model:llama-4-maverick@current",
      "kind": "serves"
    },
    {
      "from": "provider:replicate",
      "to": "model:llama-4-scout@current",
      "kind": "serves"
    }
  ],
  "incomingEdges": [
    {
      "from": "model:llama-3-1-405b-instruct@current",
      "to": "provider:replicate",
      "kind": "served_by"
    },
    {
      "from": "model:llama-3-1-70b-instruct@current",
      "to": "provider:replicate",
      "kind": "served_by"
    },
    {
      "from": "model:llama-3-3-70b-instruct@current",
      "to": "provider:replicate",
      "kind": "served_by"
    },
    {
      "from": "model:llama-4-maverick@current",
      "to": "provider:replicate",
      "kind": "served_by"
    },
    {
      "from": "model:llama-4-scout@current",
      "to": "provider:replicate",
      "kind": "served_by"
    },
    {
      "from": "tool-server:mcp-replicate",
      "to": "provider:replicate",
      "kind": "integrates_with",
      "attributes": {}
    }
  ]
}