iiRecord
Agentic AI Atlas · Ollama
provider:ollamaa5c.ai
II.
Provider JSON

provider:ollama

Structured · live

Ollama json

Inspect the normalized record payload exactly as the atlas UI reads it.

File · compute/providers/ollama.yamlCluster · compute
Record JSON
{
  "id": "provider:ollama",
  "_kind": "Provider",
  "_file": "compute/providers/ollama.yaml",
  "_cluster": "compute",
  "attributes": {
    "displayName": "Ollama",
    "vendor": "Ollama (community)",
    "versionRange": ">=0.1.0",
    "authMethods": [
      "api-key"
    ],
    "authMethodNotes": "Local-first server (default `http://localhost:11434`). No auth on the\nbare local server; `api-key` is used when fronting Ollama via a proxy\n(e.g. cloud Ollama deployments). The `api-key` enum value is selected\nhere as the closest auth-method match.\n",
    "endpoints": {
      "base": "http://localhost:11434",
      "chat": "http://localhost:11434/api/chat",
      "generate": "http://localhost:11434/api/generate",
      "embed": "http://localhost:11434/api/embed",
      "tags": "http://localhost:11434/api/tags",
      "show": "http://localhost:11434/api/show"
    },
    "pricing": "Free for local execution. Cloud-hosted Ollama deployments price\nindependently.\n",
    "rateLimitSignalingProtocol": "None on the local server. HTTP 5xx surfaces upstream model-runtime\nerrors. JSON error envelope: `{ \"error\": \"...\" }`.\n",
    "dataResidencyOptions": [
      "on-prem"
    ],
    "vendorFeatures": [],
    "slaTier": "ollama-no-sla",
    "regions": [
      "on-prem"
    ]
  },
  "outgoingEdges": [
    {
      "from": "provider:ollama",
      "to": "layer:2-provider",
      "kind": "realizes",
      "attributes": {}
    },
    {
      "from": "provider:ollama",
      "to": "model:llama-3-3-70b-instruct@current",
      "kind": "serves"
    },
    {
      "from": "provider:ollama",
      "to": "model:gemma-2-27b@current",
      "kind": "serves"
    },
    {
      "from": "provider:ollama",
      "to": "protocol-message:ollama-chat-message",
      "kind": "emits_message_type"
    },
    {
      "from": "provider:ollama",
      "to": "protocol-message:ollama-chat-stream-chunk",
      "kind": "emits_message_type"
    },
    {
      "from": "provider:ollama",
      "to": "protocol-message:ollama-chat-stream-done",
      "kind": "emits_message_type"
    },
    {
      "from": "provider:ollama",
      "to": "protocol-message:ollama-generate-response",
      "kind": "emits_message_type"
    },
    {
      "from": "provider:ollama",
      "to": "protocol-message:ollama-generate-stream-chunk",
      "kind": "emits_message_type"
    },
    {
      "from": "provider:ollama",
      "to": "protocol-message:ollama-embed-response",
      "kind": "emits_message_type"
    }
  ],
  "incomingEdges": [
    {
      "from": "claim:ollama-provider-native-api-endpoints",
      "to": "provider:ollama",
      "kind": "about_subject"
    },
    {
      "from": "claim:ollama-native-protocol-messages",
      "to": "provider:ollama",
      "kind": "about_subject"
    },
    {
      "from": "tool-server:mcp-ollama",
      "to": "provider:ollama",
      "kind": "integrates_with",
      "attributes": {}
    }
  ]
}