adaline-providers

Adaline Providers

Concepts

Adaline Providers are LLM services configured in your Adaline workspace. Each provider represents a connection to an AI service (OpenAI, Anthropic, Google, Azure, Bedrock, Vertex, Groq, xAI, OpenRouter, Together AI, or a custom endpoint). Models are specific LLMs available through each configured provider.

Key terms:

• Provider — a configured connection to an LLM service, including credentials and settings. Identified by a unique id and associated settingId.
• settingId — the internal ID for the provider's credential/configuration record.
• modelSettings — per-provider flags indicating which model parameters (temperature, maxTokens, etc.) are supported and enabled.
• Model — a specific LLM available through a provider (e.g., gpt-4o via OpenAI). Has an enabled flag controlling availability.
• providerId — the ID of the provider record; used to filter models to a specific provider.

Supported Providers

Provider name	Description
openai	OpenAI GPT models
anthropic	Anthropic Claude models
google	Google Gemini models
xai	xAI Grok models
azure	Azure OpenAI Service
bedrock	AWS Bedrock (multiple model families)
vertex	Google Vertex AI
groq	Groq ultra-low-latency inference
open-router	OpenRouter multi-provider gateway
togetherai	Together AI open-source models
custom	Custom or self-hosted LLM endpoints

Configuration

Set these environment variables when your Adaline credentials are available:

• ADALINE_API_KEY — your workspace API key (from Settings > API Keys at app.adaline.ai)
• Base URL: https://api.adaline.ai/v2

You can start integrating before you have credentials. All code examples use placeholder values — replace them with real values when ready.

How It Works

1. Configure providers in the Adaline workspace settings by supplying credentials for each LLM service
2. Each configured provider gets a unique id and exposes its available models
3. When creating or running prompts, select a provider and model from the configured list
4. Use modelSettings to understand which parameters (temperature, maxTokens, etc.) each provider supports

Checking available providers and models before creating prompts ensures you reference valid provider/model combinations and understand which model parameters are supported.

Quick Triage

Symptom	First Fix
Provider not in list	Verify credentials are configured in Adaline workspace settings
Model not found for provider	Use GET /models?providerId=ID to list models for that specific provider
Model shows enabled: false	Enable the model in Adaline workspace settings
modelSettings field missing a param	That provider does not support that parameter — omit it from prompt config

List All Providers

curl -X GET "https://api.adaline.ai/v2/providers" \
  -H "Authorization: Bearer $ADALINE_API_KEY"

Response:

{
  "providers": [
    {
      "id": "prov_abc123",
      "settingId": "set_xyz789",
      "name": "openai",
      "description": "OpenAI GPT models",
      "modelSettings": {
        "temperature": { "enabled": true },
        "maxTokens": { "enabled": true }
      }
    },
    {
      "id": "prov_def456",
      "settingId": "set_uvw012",
      "name": "anthropic",
      "description": "Anthropic Claude models",
      "modelSettings": {
        "temperature": { "enabled": true },
        "maxTokens": { "enabled": true }
      }
    }
  ]
}

Get a Specific Provider

curl -X GET "https://api.adaline.ai/v2/providers/prov_abc123" \
  -H "Authorization: Bearer $ADALINE_API_KEY"

Response:

{
  "id": "prov_abc123",
  "settingId": "set_xyz789",
  "name": "openai",
  "description": "OpenAI GPT models",
  "modelSettings": {
    "temperature": { "enabled": true },
    "maxTokens": { "enabled": true }
  }
}

List Available Models

# All models across all providers
curl -X GET "https://api.adaline.ai/v2/models" \
  -H "Authorization: Bearer $ADALINE_API_KEY"

# Models for a specific provider
curl -X GET "https://api.adaline.ai/v2/models?providerId=prov_abc123" \
  -H "Authorization: Bearer $ADALINE_API_KEY"

Response:

{
  "models": [
    {
      "id": "mdl_111aaa",
      "providerId": "prov_abc123",
      "provider": "openai",
      "name": "gpt-4o",
      "enabled": true
    },
    {
      "id": "mdl_222bbb",
      "providerId": "prov_abc123",
      "provider": "openai",
      "name": "gpt-4o-mini",
      "enabled": true
    }
  ]
}

Best Practices

1. Call GET /providers at startup to discover which LLM services are configured — do not hardcode provider IDs
2. Call GET /models?providerId=ID before creating a prompt to confirm the target model is available and enabled
3. Check modelSettings to determine which parameters a provider supports before including them in prompt configuration — unsupported parameters are ignored or may cause errors
4. Cache provider and model lists at application startup; they change infrequently
5. Use the enabled flag on models to filter out unavailable models before presenting choices to users

References

See references/api.md for the full REST API reference with request parameters, response schemas, and curl examples.