AI Models

​

Overview

Draft & Goal supports multiple AI model providers, giving you flexibility to choose the best models for your workflows. You can enable built-in providers or add custom providers by configuring API keys. Available providers:

• Built-in providers: Google Vertex AI, AWS Bedrock (pre-configured)
• Custom providers: OpenAI, Anthropic, Google Gemini, OVHcloud (requires API key)

​

Accessing AI Models

Navigate to Settings → Models in the studio to view and manage your AI model providers. The Models page displays:

• All available model providers
• Individual models within each provider
• Activation status for each model
• Configuration options for custom providers

​

Built-in providers

​

Google Vertex AI

Google Vertex AI models are pre-configured and available by default:

• claude-opus-4-1@20250805
• claude-sonnet-4-5@20250929
• claude-sonnet-4@20250514
• gemini-2.5-flash
• gemini-2.5-flash-image
• gemini-2.5-pro
• gemini-3-pro-preview
• imagen-4.0-generate-001
• veo-3.0-generate-001

​

AWS Bedrock

AWS Bedrock models are also pre-configured and ready to use. Enable the models your team needs from the list.

​

Enabling and disabling models

Each model can be individually enabled or disabled:

1. Navigate to the provider section
2. Click the toggle switch next to each model
3. Enabled models appear in blue
4. Changes are saved automatically

​

Adding custom providers

Add additional model providers by configuring their API keys.

​

OpenAI

To use OpenAI models (GPT-4, GPT-3.5, etc.):

1. Click Configure on the OpenAI card
2. Enter your OpenAI API key
3. Click Save

Getting an OpenAI API key:

1. Go to platform.openai.com
2. Navigate to API Keys
3. Create a new key
4. Copy and paste into Draft & Goal

​

Anthropic

To use Anthropic models (Claude):

1. Click Configure on the Anthropic card
2. Enter your Anthropic API key
3. Click Save

Getting an Anthropic API key:

1. Go to console.anthropic.com
2. Navigate to API Keys
3. Create a new key
4. Copy and paste into Draft & Goal

​

Google Gemini

To use Google Gemini models via Google AI Studio:

1. Click Configure on the Gemini card
2. Enter your Google AI API key
3. Click Save

Getting a Google AI API key:

1. Go to makersuite.google.com
2. Navigate to Get API Key
3. Create a new key
4. Copy and paste into Draft & Goal

​

OVHcloud

To use OVHcloud AI models:

1. Click Configure on the OVH card
2. Enter your OVHcloud API key
3. Click Save

Getting an OVHcloud API key:

1. Go to your OVHcloud console
2. Navigate to AI Services
3. Generate an API key
4. Copy and paste into Draft & Goal

​

Using models in workflows

Once models are enabled, they become available in AI nodes:

1. Add an LLM or AI Agent node to your workflow
2. Open the node settings
3. In the Model dropdown, select from available models
4. Models are organized by provider

The model selector shows:

• Provider name (e.g., OPENAI, GOOGLE_VERTEXAI, ANTHROPIC)
• Model name (e.g., gpt-4o, claude-sonnet-4-5)
• Only enabled models appear in the list

​

Best practices

​

Security

• Protect API keys: Never share API keys in workflow descriptions or public documentation
• Use workspace keys: Configure keys at the workspace level so individual users don’t need their own
• Rotate regularly: Update API keys periodically for security
• Monitor usage: Track API usage to detect unauthorized access

​

Cost management

• Start small: Begin with faster, cheaper models for testing
• Enable selectively: Only enable models your team actually uses
• Monitor spend: Most providers offer usage dashboards
• Set limits: Configure spending limits with your provider

​

Model selection

• Match task to model: Use simpler models for simple tasks
• Test different models: Compare results across providers
• Consider latency: Faster models improve workflow execution time
• Check availability: Ensure models are available in your region

​

Organization

​

Troubleshooting

​

API key rejected

​

Model not appearing

If a model doesn’t appear in workflow nodes:

1. Verify the model is enabled (toggle is blue)
2. Check that the provider is configured correctly
3. Refresh the workflow editor
4. Clear browser cache if needed

​

Rate limits

If you encounter rate limit errors:

1. Check your provider’s rate limits
2. Upgrade your provider plan if needed
3. Implement retry logic in workflows
4. Distribute requests across multiple models

​

Connection errors

If you can’t connect to a provider:

1. Verify your API key is correct
2. Check your internet connection
3. Ensure the provider service is operational
4. Contact the provider’s support if issues persist

​

Provider comparison

Different providers have different strengths:

​

Next steps