Skip to content
AlphaBase
HomeDocs

Providers

Provider auth, BYOK cost model, filtering, and model override contracts.

Providers are model backends AlphaBase can route through. Provider metadata is sourced from Models.dev and then merged with your connected credentials and local overrides.

BYOK Cost Model

AlphaBase does not bundle provider credits or promise a default free model. In the usual BYOK flow, you connect a provider with an API key or OAuth and that provider applies its own pricing, quotas, and data policies. AlphaBase does not add a markup to provider usage.

You can also configure a local runtime such as Ollama, LM Studio, or another OpenAI-compatible endpoint. Local execution may avoid hosted API charges, but it still requires a working runtime, model files, and local compute. An Ollama model whose identifier ends in :cloud is a cloud service path and still depends on Ollama access and its quota; it is not a bundled AlphaBase entitlement. See Models for selection and reliability details.

Runtime Provider APIs

Useful server routes:

  • GET /provider
  • GET /provider/auth
  • GET /provider/capability-index
  • POST /provider/:providerID/oauth/authorize
  • POST /provider/:providerID/oauth/callback
  • GET /provider/model-search?q=<query>

Connect A Provider

Typical flow:

  1. open provider UI or /connect flow
  2. authenticate via OAuth or API key
  3. confirm provider appears in connected list
  4. choose a model

Filter Provider Inventory

Use config-level filters:

{
  "enabled_providers": ["openai", "anthropic", "ollama"],
  "disabled_providers": ["openrouter"]
}

enabled_providers is an allowlist. If present, all non-listed providers are hidden.

Provider Override Schema

Each provider.<id> can override provider or model behavior.

{
  "provider": {
    "openai": {
      "options": {
        "apiKey": "{env:OPENAI_API_KEY}",
        "timeout": 300000,
        "setCacheKey": true,
      },
      "models": {
        "<model-id>": {
          "options": {
            "reasoningEffort": "high",
          },
          "variants": {
            "fast": { "disabled": true },
          },
        },
      },
    },
  },
}

Supported common option fields:

  • apiKey
  • baseURL
  • enterpriseUrl (notably for GitHub Enterprise Copilot auth)
  • setCacheKey
  • timeout (number or false)

Custom Provider Pattern

You can define non-default providers via provider records (for example OpenAI-compatible gateways), then route models through them.

Model Search (Model-First Onboarding)

/provider/model-search groups matching models by canonical family and ranks provider options by first-party/bundled/connected status. This powers “add model first, then choose provider” workflows.

Troubleshooting

If a provider is configured but unusable:

  1. check runtime/capabilities
  2. verify env key scope and endpoint URL
  3. verify provider is not excluded by enabled_providers / disabled_providers
  4. confirm timeout/baseURL overrides are valid
  5. if the model intermittently fails, check whether it has been classified degraded from recent evidence — see Models; this self-heals as soon as the model’s recent attempts improve.

See Troubleshooting.