Logo Mimicry Docs

AI Providers

Mimicry is built to be model-agnostic. Configure multiple AI providers to ensure high availability and cost-efficiency.

Supported Providers

OpenAI

Native support for GPT-4o, GPT-4, and GPT-3.5 Turbo.

OpenRouter

Access hundreds of models (Claude, Llama, Mistral) via a single API.

Gemini

Google's powerful generative AI models with high rate limits.

DeepSeek

Cost-effective reasoning and coding models.

Configuring a Provider

To add a new AI provider, navigate to Settings > AI Providers in the admin dashboard.

settings Configuration Fields

  • Provider Name: A unique label for this provider (e.g., "Main OpenAI").
  • Base URL: The API endpoint (e.g., https://api.openai.com/v1).
  • API Key: Your private authentication token.
  • Model Selection: Choose the preferred model for this provider.
  • Max Tokens: Limit the length of generated content to control costs.
  • Temperature: Control the creativity vs. consistency of the output.

Intelligent Failover

Mimicry implements a smart retry mechanism. If your primary provider experiences an outage or hits a rate limit, the system can automatically switch to a fallback provider to ensure your content generation campaigns continue uninterrupted.

Best Practices

  • API Keys: Use restricted API keys when possible for better security.
  • Cost Control: Set monthly limits on your provider dashboards.
  • Model Testing: Test different models to find the best balance between quality and cost for your specific niche.