Skip to content
Cheffed

AI Providers

AI providers supply the language models that power your Cheffed sessions. Configure one or more providers to enable session creation.

Supported Providers

Section titled “Supported Providers”
ProviderModelsRecommended For
AnthropicClaude 3.5 Sonnet, Claude 3 OpusComplex coding tasks, reasoning
OpenAIGPT-4, GPT-4 TurboGeneral tasks, broad knowledge

Anthropic

Section titled “Anthropic”

Anthropic’s Claude models excel at coding tasks with strong reasoning capabilities.

Getting an API Key

Section titled “Getting an API Key”
  1. Go to console.anthropic.com
  2. Sign up or log in to your account
  3. Navigate to API Keys
  4. Click Create Key
  5. Copy the key (it won’t be shown again)

Adding to Cheffed

Section titled “Adding to Cheffed”
  1. Go to Settings in Cheffed
  2. Find the Anthropic section
  3. Click Add Credentials
  4. Paste your API key
  5. Click Save

Available Models

Section titled “Available Models”
ModelSpeedCapabilityBest For
Claude 3.5 SonnetFastHighMost tasks (recommended)
Claude 3 OpusSlowerHighestComplex reasoning
Claude 3 HaikuFastestGoodQuick tasks

Pricing

Section titled “Pricing”

Anthropic charges per token:

  • Input tokens - Text you send to the model
  • Output tokens - Text the model generates

Check anthropic.com/pricing for current rates.

OpenAI

Section titled “OpenAI”

OpenAI’s GPT models offer broad capabilities and general knowledge.

Getting an API Key

Section titled “Getting an API Key”
  1. Go to platform.openai.com
  2. Sign up or log in to your account
  3. Navigate to API Keys
  4. Click Create new secret key
  5. Copy the key (it won’t be shown again)

Adding to Cheffed

Section titled “Adding to Cheffed”
  1. Go to Settings in Cheffed
  2. Find the OpenAI section
  3. Click Add Credentials
  4. Paste your API key
  5. Click Save

Available Models

Section titled “Available Models”
ModelSpeedCapabilityBest For
GPT-4SlowerHighestComplex tasks
GPT-4 TurboFastHighMost tasks
GPT-4oFastestHighQuick tasks

Pricing

Section titled “Pricing”

OpenAI charges per token with different rates per model. Check openai.com/pricing for current rates.

Choosing a Provider

Section titled “Choosing a Provider”

For Most Users

Section titled “For Most Users”

Start with Anthropic Claude 3.5 Sonnet:

  • Excellent coding capabilities
  • Fast responses
  • Competitive pricing
  • Strong reasoning

For Specific Needs

Section titled “For Specific Needs”
NeedRecommendation
Complex reasoningClaude 3 Opus
Fast responsesClaude 3 Haiku or GPT-4o
Broad knowledgeGPT-4
Budget consciousClaude 3.5 Sonnet

Using Multiple Providers

Section titled “Using Multiple Providers”

You can configure multiple providers and choose per-session:

  1. Add credentials for each provider
  2. When creating a session, select your preferred model
  3. Different automations can use different models

Benefits of Multiple Providers

Section titled “Benefits of Multiple Providers”
  • Redundancy - Fallback if one provider has issues
  • Optimization - Use best model for each task type
  • Cost management - Balance quality vs. cost

Credential Management

Section titled “Credential Management”

Viewing Credentials

Section titled “Viewing Credentials”

Go to Settings to see all configured providers:

  • Provider name
  • Status (connected/disconnected)
  • Last used date

Updating Credentials

Section titled “Updating Credentials”

To update an API key:

  1. Go to Settings
  2. Find the provider
  3. Click Edit
  4. Enter the new key
  5. Save changes

Removing Credentials

Section titled “Removing Credentials”

To remove a provider:

  1. Go to Settings
  2. Find the provider
  3. Click Remove
  4. Confirm deletion

Troubleshooting

Section titled “Troubleshooting”

Invalid API Key

Section titled “Invalid API Key”
SymptomSolution
”Invalid API key” errorVerify key is correct, no extra spaces
”Authentication failed”Generate a new key from provider
Key not savingCheck for special characters, try new key

Rate Limits

Section titled “Rate Limits”

Providers enforce rate limits:

ProviderTypical Limits
AnthropicVaries by tier
OpenAIVaries by tier

If you hit rate limits:

  1. Wait and retry
  2. Upgrade your provider plan
  3. Use a different provider temporarily

Billing Issues

Section titled “Billing Issues”

Sessions fail if provider billing is not set up:

  1. Log into provider console
  2. Add payment method
  3. Ensure account is active
  4. Retry session

Security

Section titled “Security”

Key Storage

Section titled “Key Storage”
  • API keys are encrypted at rest
  • Keys are never exposed in the UI after saving
  • Keys are only used server-side

Best Practices

Section titled “Best Practices”
  1. Use dedicated API keys for Cheffed
  2. Set spending limits at the provider
  3. Monitor usage regularly
  4. Rotate keys periodically

Cost Management

Section titled “Cost Management”

Monitoring Usage

Section titled “Monitoring Usage”

Track your AI costs:

  • View provider dashboards for usage
  • Monitor session count in Cheffed
  • Set up provider spending alerts

Reducing Costs

Section titled “Reducing Costs”
  1. Choose appropriate models for tasks
  2. Use faster/cheaper models for simple tasks
  3. Write clear prompts to reduce back-and-forth
  4. Set reasonable session timeouts

Next Steps

Section titled “Next Steps”