Skip to content
lobu
API Blog

Providers

Lobu separates inference from execution.

  • An inference provider supplies models.
  • An environment selects where commands run.
  • A runtime provider implements that sandbox backend.
  • Credentials stay gateway-side.

See Concepts and architecture for the complete agent model.

Provider terms

Section titled “Provider terms”
TermMeaning
Provider catalog entryBuilt-in metadata from config/providers.json
Inference providerA configured credential, optionally with organization-specific upstreams and defaults
ModelThe text, image, speech, or embedding model selected for a run
Auth profileA reusable OAuth, device-code, or API-key credential
EnvironmentA named execution target selected by an agent
Runtime providerThe implementation behind a remote sandbox environment

Inference and sandbox compute are independent layers.

Model selection

Section titled “Model selection”

The text model resolves in this order:

  1. automation or run override
  2. agent default model
  3. organization default inference provider

This lets one agent use a cheap model for routine watches and a stronger model for a sensitive workflow without changing its identity or environment.

Organization-defined inference

Section titled “Organization-defined inference”

An organization-defined provider has a stable slug, a provider kind, one gateway-held credential, and optional per-modality configuration.

Supported modality blocks are:

  • text
  • image
  • stt
  • tts
  • embedding

Each block may define its own base URL, model, and model-discovery endpoint.

Credentials

Section titled “Credentials”

Credential sources include API keys, OAuth, device-code authentication, declared project secrets, user-scoped auth profiles, and organization-owned provider credentials.

The gateway resolves the selected credential. Workers and remote sandboxes receive scoped access or placeholders rather than durable raw secrets.

Inference is not the sandbox

Section titled “Inference is not the sandbox”

An agent also selects an environment:

EnvironmentExecution
BuiltinLocal embedded runtime
DeviceAn authorized user device
Provider-backedA named organization or private remote sandbox

A runtime provider handles sandbox creation, workspace mapping, network policy, and command execution. The same agent can use an Anthropic model while executing in a Vercel-backed environment, or use an organization-defined OpenAI-compatible endpoint while running on a device.

Configure an agent

Section titled “Configure an agent”
const support = defineAgent({
  id: "support",
  providers: [
    {
      id: "openrouter",
      model: "anthropic/claude-sonnet-4",
      key: secret("OPENROUTER_API_KEY"),
    },
  ],
  // environment selection is separate
});

Use the lobu.config.ts reference for the exact schema. Organization-defined providers and environments can also be managed through the UI and API.

Built-in provider registry

Section titled “Built-in provider registry”
ProviderIDDefault ModelSDK CompatBase URLModels Endpoint
Claudeclaudeclaude-sonnet-5anthropichttps://api.anthropic.comN/A
ChatGPTchatgptN/AN/Ahttps://chatgpt.com/backend-apiN/A
Groqgroqllama-3.3-70b-versatileopenaihttps://api.groq.com/openai/v1/models
Geminigeminigemini-2.5-proopenaihttps://generativelanguage.googleapis.com/v1beta/openaiN/A
Together AItogether-aimeta-llama/Llama-3.3-70B-Instruct-Turboopenaihttps://api.together.xyz/v1/models
NVIDIA NIM (free)nvidianvidia/moonshotai/kimi-k2.6openaihttps://integrate.api.nvidia.com/v1/models
Fireworks AIfireworksaccounts/fireworks/models/llama-v3p3-70b-instructopenaihttps://api.fireworks.ai/inference/v1/models
Mistralmistralmistral-large-latestopenaihttps://api.mistral.ai/v1/models
DeepSeekdeepseekdeepseek-v4-flashopenaihttps://api.deepseek.com/v1/models
OpenRouteropenrouteranthropic/claude-sonnet-5openaihttps://openrouter.ai/api/v1/models
Cerebrascerebrasllama-3.3-70bopenaihttps://api.cerebras.ai/v1/models
OpenCode Zenopencode-zenclaude-sonnet-4-6openaihttps://opencode.ai/zen/v1/models
Qwenqwenqwen-maxopenaihttps://dashscope.aliyuncs.com/compatible-mode/v1/models
xAI (SuperGrok)xaigrok-4openaihttps://api.x.ai/v1/models
Perplexityperplexitysonaropenaihttps://api.perplexity.ai/v1/models
Coherecoherecommand-a-03-2025openaihttps://api.cohere.com/compatibility/v1/models
OpenAI APIopenaigpt-5.6-solopenaihttps://api.openai.com/v1/models

Source of truth

Section titled “Source of truth”
View provider entries
[
  {
    "id": "claude",
    "providers": [
      {
        "displayName": "Claude",
        "iconUrl": "https://www.google.com/s2/favicons?domain=anthropic.com&sz=128",
        "envVarName": "ANTHROPIC_API_KEY",
        "upstreamBaseUrl": "https://api.anthropic.com",
        "apiKeyInstructions": "Get your API key from <a href=\"https://console.anthropic.com/settings/keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">Anthropic Console</a>, or sign in with Claude",
        "apiKeyPlaceholder": "sk-ant-...",
        "sdkCompat": "anthropic",
        "modalities": [
          "text"
        ],
        "defaultModel": "claude-sonnet-5",
        "models": [
          "claude-opus-4-8",
          "claude-sonnet-5",
          "claude-haiku-4-5-20251001",
          "claude-fable-5"
        ],
        "oauth": {
          "clientId": "9d1c250a-e61b-44d9-88ed-5944d1962f5e",
          "authUrl": "https://claude.com/cai/oauth/authorize",
          "tokenUrl": "https://platform.claude.com/v1/oauth/token",
          "redirectUri": "https://platform.claude.com/oauth/code/callback",
          "scope": "user:inference",
          "responseType": "code",
          "grantType": "authorization_code",
          "tokenEndpointAuthMethod": "none",
          "extraAuthParams": {
            "code": "true"
          },
          "tokenRequestFormat": "form",
          "grant": "authorization-code",
          "authType": "oauth"
        }
      }
    ]
  },
  {
    "id": "chatgpt",
    "providers": [
      {
        "displayName": "ChatGPT",
        "iconUrl": "https://www.google.com/s2/favicons?domain=chatgpt.com&sz=128",
        "envVarName": "OPENAI_API_KEY",
        "upstreamBaseUrl": "https://chatgpt.com/backend-api",
        "apiKeyInstructions": "Sign in with ChatGPT Plus/Pro, or paste an OpenAI API key",
        "apiKeyPlaceholder": "sk-...",
        "modalities": [
          "text"
        ],
        "oauth": {
          "clientId": "app_EMoamEEZ73f0CkXaXp7hrann",
          "tokenUrl": "https://auth.openai.com/oauth/token",
          "redirectUri": "https://auth.openai.com/deviceauth/callback",
          "scope": "openid profile email offline_access api.model.read api.model.request api.model.image.request api.model.audio.request",
          "tokenEndpointAuthMethod": "none",
          "tokenRequestFormat": "form",
          "grant": "openai-device-auth",
          "authType": "device-code",
          "deviceCodeUrl": "https://auth.openai.com/api/accounts/deviceauth/usercode",
          "deviceTokenUrl": "https://auth.openai.com/api/accounts/deviceauth/token",
          "deviceRedirectUri": "https://auth.openai.com/deviceauth/callback",
          "defaultVerificationUrl": "https://auth.openai.com/codex/device",
          "includeScopeInRefresh": true,
          "accountIdClaimPath": "https://api.openai.com/auth",
          "scopeEnvVar": "OPENAI_OAUTH_SCOPE",
          "userAgent": "reqwest/0.12.24"
        }
      }
    ]
  },
  {
    "id": "groq",
    "providers": [
      {
        "displayName": "Groq",
        "iconUrl": "https://www.google.com/s2/favicons?domain=groq.com&sz=128",
        "envVarName": "GROQ_API_KEY",
        "upstreamBaseUrl": "https://api.groq.com/openai/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://console.groq.com/keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">Groq Console</a>",
        "apiKeyPlaceholder": "gsk_...",
        "sdkCompat": "openai",
        "modalities": [
          "text",
          "stt"
        ],
        "defaultModel": "llama-3.3-70b-versatile",
        "modelsEndpoint": "/models",
        "stt": {
          "enabled": true,
          "sdkCompat": "openai",
          "transcriptionPath": "/audio/transcriptions",
          "model": "whisper-large-v3-turbo"
        },
        "models": [
          "llama-3.3-70b-versatile",
          "llama-3.1-8b-instant",
          "mixtral-8x7b-32768"
        ]
      }
    ]
  },
  {
    "id": "gemini",
    "providers": [
      {
        "displayName": "Gemini",
        "iconUrl": "https://www.google.com/s2/favicons?domain=gemini.google.com&sz=128",
        "envVarName": "GEMINI_API_KEY",
        "upstreamBaseUrl": "https://generativelanguage.googleapis.com/v1beta/openai",
        "apiKeyInstructions": "Get your API key from <a href=\"https://aistudio.google.com/apikey\" target=\"_blank\" class=\"text-blue-600 hover:underline\">Google AI Studio</a>",
        "apiKeyPlaceholder": "AIza...",
        "sdkCompat": "openai",
        "modalities": [
          "text",
          "image"
        ],
        "defaultModel": "gemini-2.5-pro",
        "models": [
          "gemini-2.5-pro",
          "gemini-2.5-flash",
          "gemini-2.0-flash"
        ]
      }
    ]
  },
  {
    "id": "together-ai",
    "providers": [
      {
        "displayName": "Together AI",
        "iconUrl": "https://www.google.com/s2/favicons?domain=together.ai&sz=128",
        "envVarName": "TOGETHER_API_KEY",
        "upstreamBaseUrl": "https://api.together.xyz/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://api.together.ai/settings/api-keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">Together AI Dashboard</a>",
        "apiKeyPlaceholder": "tok_...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "meta-llama/Llama-3.3-70B-Instruct-Turbo",
        "modelsEndpoint": "/models",
        "models": [
          "meta-llama/Llama-3.3-70B-Instruct-Turbo",
          "Qwen/Qwen2.5-72B-Instruct-Turbo",
          "deepseek-ai/DeepSeek-V3"
        ]
      }
    ]
  },
  {
    "id": "nvidia",
    "providers": [
      {
        "displayName": "NVIDIA NIM (free)",
        "iconUrl": "https://www.google.com/s2/favicons?domain=nvidia.com&sz=128",
        "envVarName": "NVIDIA_API_KEY",
        "upstreamBaseUrl": "https://integrate.api.nvidia.com/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://build.nvidia.com/settings/api-keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">NVIDIA Build</a>",
        "apiKeyPlaceholder": "nvapi-...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "modelsEndpoint": "/models",
        "defaultModel": "nvidia/moonshotai/kimi-k2.6",
        "models": [
          "nvidia/moonshotai/kimi-k2.6",
          "meta/llama-3.3-70b-instruct",
          "deepseek-ai/deepseek-r1"
        ]
      }
    ]
  },
  {
    "id": "fireworks",
    "providers": [
      {
        "displayName": "Fireworks AI",
        "iconUrl": "https://www.google.com/s2/favicons?domain=fireworks.ai&sz=128",
        "envVarName": "FIREWORKS_API_KEY",
        "upstreamBaseUrl": "https://api.fireworks.ai/inference/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://fireworks.ai/account/api-keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">Fireworks Dashboard</a>",
        "apiKeyPlaceholder": "fw_...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "accounts/fireworks/models/llama-v3p3-70b-instruct",
        "modelsEndpoint": "/models",
        "models": [
          "accounts/fireworks/models/llama-v3p3-70b-instruct",
          "accounts/fireworks/models/deepseek-v3",
          "accounts/fireworks/models/qwen2p5-72b-instruct"
        ]
      }
    ]
  },
  {
    "id": "mistral",
    "providers": [
      {
        "displayName": "Mistral",
        "iconUrl": "https://www.google.com/s2/favicons?domain=mistral.ai&sz=128",
        "envVarName": "MISTRAL_API_KEY",
        "upstreamBaseUrl": "https://api.mistral.ai/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://console.mistral.ai/api-keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">Mistral Console</a>",
        "apiKeyPlaceholder": "sk-...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "mistral-large-latest",
        "modelsEndpoint": "/models",
        "models": [
          "mistral-large-latest",
          "mistral-small-latest",
          "codestral-latest"
        ]
      }
    ]
  },
  {
    "id": "deepseek",
    "providers": [
      {
        "displayName": "DeepSeek",
        "iconUrl": "https://www.google.com/s2/favicons?domain=deepseek.com&sz=128",
        "envVarName": "DEEPSEEK_API_KEY",
        "upstreamBaseUrl": "https://api.deepseek.com",
        "apiKeyInstructions": "Get your API key from <a href=\"https://platform.deepseek.com/api_keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">DeepSeek Platform</a>",
        "apiKeyPlaceholder": "sk-...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "deepseek-v4-flash",
        "modelsEndpoint": "/v1/models",
        "models": [
          "deepseek-v4-flash",
          "deepseek-chat",
          "deepseek-reasoner"
        ]
      }
    ]
  },
  {
    "id": "openrouter",
    "providers": [
      {
        "displayName": "OpenRouter",
        "iconUrl": "https://www.google.com/s2/favicons?domain=openrouter.ai&sz=128",
        "envVarName": "OPENROUTER_API_KEY",
        "upstreamBaseUrl": "https://openrouter.ai/api/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://openrouter.ai/keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">OpenRouter</a>, or connect via OAuth for per-user billing",
        "apiKeyPlaceholder": "sk-or-...",
        "sdkCompat": "openai",
        "modalities": [
          "text",
          "stt"
        ],
        "defaultModel": "anthropic/claude-sonnet-5",
        "modelsEndpoint": "/models",
        "stt": {
          "enabled": true,
          "sdkCompat": "openai",
          "transcriptionPath": "/audio/transcriptions",
          "model": "whisper-1"
        },
        "models": [
          "anthropic/claude-sonnet-5",
          "openai/gpt-5.6-sol",
          "google/gemini-2.5-pro"
        ]
      }
    ]
  },
  {
    "id": "cerebras",
    "providers": [
      {
        "displayName": "Cerebras",
        "iconUrl": "https://www.google.com/s2/favicons?domain=cerebras.ai&sz=128",
        "envVarName": "CEREBRAS_API_KEY",
        "upstreamBaseUrl": "https://api.cerebras.ai/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://cloud.cerebras.ai/\" target=\"_blank\" class=\"text-blue-600 hover:underline\">Cerebras Cloud</a>",
        "apiKeyPlaceholder": "csk-...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "llama-3.3-70b",
        "modelsEndpoint": "/models",
        "models": [
          "llama-3.3-70b",
          "llama3.1-8b",
          "qwen-3-32b"
        ]
      }
    ]
  },
  {
    "id": "opencode-zen",
    "providers": [
      {
        "displayName": "OpenCode Zen",
        "iconUrl": "https://www.google.com/s2/favicons?domain=opencode.ai&sz=128",
        "envVarName": "OPENCODE_ZEN_API_KEY",
        "upstreamBaseUrl": "https://opencode.ai/zen/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://opencode.ai/auth\" target=\"_blank\" class=\"text-blue-600 hover:underline\">OpenCode Zen</a>",
        "apiKeyPlaceholder": "zen-...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "claude-sonnet-4-6",
        "modelsEndpoint": "/models"
      }
    ]
  },
  {
    "id": "qwen",
    "providers": [
      {
        "displayName": "Qwen",
        "iconUrl": "https://www.google.com/s2/favicons?domain=alibabacloud.com&sz=128",
        "envVarName": "DASHSCOPE_API_KEY",
        "upstreamBaseUrl": "https://dashscope.aliyuncs.com/compatible-mode/v1",
        "apiKeyInstructions": "Get your API key from the <a href=\"https://dashscope.console.aliyun.com/apiKey\" target=\"_blank\" class=\"text-blue-600 hover:underline\">DashScope Console</a>. A Qoder / Token Plan subscription is served from a different host — set a custom upstream of <code>https://token-plan.ap-southeast-1.maas.aliyuncs.com/compatible-mode/v1</code> and paste that plan's key. The <code>qwen3.8-max</code> tiers are only reachable that way.",
        "apiKeyPlaceholder": "sk-...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "qwen-max",
        "modelsEndpoint": "/models",
        "models": [
          "qwen3.8-max",
          "qwen3.8-max-preview",
          "qwen-max",
          "qwen-plus",
          "qwen-turbo",
          "qwen2.5-coder-32b-instruct",
          "qwen2.5-72b-instruct"
        ]
      }
    ]
  },
  {
    "id": "xai",
    "providers": [
      {
        "displayName": "xAI (SuperGrok)",
        "iconUrl": "https://www.google.com/s2/favicons?domain=x.ai&sz=128",
        "envVarName": "XAI_API_KEY",
        "upstreamBaseUrl": "https://api.x.ai/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://console.x.ai/\" target=\"_blank\" class=\"text-blue-600 hover:underline\">xAI Console</a>, or sign in with SuperGrok / X Premium+",
        "apiKeyPlaceholder": "xai-...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "grok-4",
        "modelsEndpoint": "/models",
        "models": [
          "grok-4",
          "grok-3",
          "grok-3-mini"
        ],
        "oauth": {
          "clientId": "b1a00492-073a-47ea-816f-4c329264a828",
          "tokenUrl": "https://auth.x.ai/oauth2/token",
          "scope": "openid profile email offline_access grok-cli:access api:access",
          "tokenEndpointAuthMethod": "none",
          "tokenRequestFormat": "form",
          "grant": "device-code",
          "authType": "device-code",
          "deviceCodeUrl": "https://auth.x.ai/oauth2/device/code",
          "defaultVerificationUrl": "https://accounts.x.ai/oauth2/device",
          "verificationUserCodeParam": "user_code"
        }
      }
    ]
  },
  {
    "id": "perplexity",
    "providers": [
      {
        "displayName": "Perplexity",
        "iconUrl": "https://www.google.com/s2/favicons?domain=perplexity.ai&sz=128",
        "envVarName": "PERPLEXITY_API_KEY",
        "upstreamBaseUrl": "https://api.perplexity.ai",
        "apiKeyInstructions": "Get your API key from <a href=\"https://www.perplexity.ai/settings/api\" target=\"_blank\" class=\"text-blue-600 hover:underline\">Perplexity Settings</a>",
        "apiKeyPlaceholder": "pplx-...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "sonar",
        "modelsEndpoint": "/v1/models",
        "models": [
          "sonar",
          "sonar-pro",
          "sonar-reasoning-pro"
        ]
      }
    ]
  },
  {
    "id": "cohere",
    "providers": [
      {
        "displayName": "Cohere",
        "iconUrl": "https://www.google.com/s2/favicons?domain=cohere.com&sz=128",
        "envVarName": "COHERE_API_KEY",
        "upstreamBaseUrl": "https://api.cohere.com/compatibility/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://dashboard.cohere.com/api-keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">Cohere Dashboard</a>",
        "apiKeyPlaceholder": "co-...",
        "sdkCompat": "openai",
        "modalities": [
          "text"
        ],
        "defaultModel": "command-a-03-2025",
        "modelsEndpoint": "/models",
        "models": [
          "command-a-03-2025",
          "command-r-plus",
          "command-r"
        ]
      }
    ]
  },
  {
    "id": "openai",
    "providers": [
      {
        "displayName": "OpenAI API",
        "iconUrl": "https://www.google.com/s2/favicons?domain=openai.com&sz=128",
        "envVarName": "OPENAI_API_KEY",
        "upstreamBaseUrl": "https://api.openai.com/v1",
        "apiKeyInstructions": "Get your API key from <a href=\"https://platform.openai.com/api-keys\" target=\"_blank\" class=\"text-blue-600 hover:underline\">OpenAI Dashboard</a>",
        "apiKeyPlaceholder": "sk-...",
        "sdkCompat": "openai",
        "modalities": [
          "text",
          "image",
          "stt",
          "tts"
        ],
        "stt": {
          "enabled": true,
          "sdkCompat": "openai",
          "transcriptionPath": "/audio/transcriptions",
          "model": "gpt-4o-transcribe"
        },
        "defaultModel": "gpt-5.6-sol",
        "modelsEndpoint": "/models",
        "models": [
          "gpt-5.6-sol",
          "gpt-5.6-terra",
          "gpt-5.6-luna",
          "gpt-5.5"
        ]
      }
    ]
  }
]
Section titled “Read next”
Start with the guideUse the reference after the core model is clear.CLI referenceScaffold, run, authenticate, and manage resources.Embed into your appCall agents and stream replies from TypeScript.