Multi Provider External Pipe

Multi-Provider LLM Integration w/ Auto-Discovery, Custom Branding, & Complete Model Support

A powerful OpenWebUI pipe that automatically discovers and integrates the latest models from OpenAI, Anthropic, and Google APIs with zero manual updates required.

✨ Key Features

• 🔄 Auto-Discovery: Automatically finds new models as they're released
• 🎨 Custom Branding: Configurable model prefixes (MPE:, External:, etc.)
• 🔓 Permissive by Default: No content filtering on Google models
• ⚡ Smart Caching: 60-minute cache for optimal performance
• 🛡️ Bulletproof Fallbacks: Hardcoded models if discovery fails
• 🔧 Fully Configurable: All settings controllable via valves

🚀 Supported Providers

• OpenAI: GPT-4o, GPT-4o Mini, o1-preview, o1-mini, GPT-3.5 Turbo + auto-discovery
• Anthropic: Claude 3.5 Sonnet/Haiku, Claude 3 Opus/Sonnet/Haiku + auto-discovery
• Google: Gemini 1.5 Pro/Flash, Gemini Pro + auto-discovery

📦 Installation

1. Copy the pipe code into OpenWebUI
2. Configure your API keys in the valves:
   • OPENAI_API_KEY
   • ANTHROPIC_API_KEY
   • GOOGLE_API_KEY
3. Customize settings as needed
4. Enjoy automatic model discovery!

⚙️ Configuration

Setting	Default	Description
MODEL_PREFIX	"MPE:"	Custom prefix for model names
AUTO_DISCOVER_MODELS	true	Enable automatic model discovery
CACHE_DURATION_MINUTES	60	How long to cache discovered models
USE_PERMISSIVE_SAFETY	true	Disable Google content filtering

🎯 Why Use This Pipe?

• Stay Current: Never miss new model releases again
• Zero Maintenance: No manual updates required
• Maximum Flexibility: Use all major LLM providers in one place
• Professional: Clean, branded model names in your dropdown

🔧 Requirements

pydantic>=2.0.0 requests>=2.0.0 google-generativeai>=0.3.0

🌟 Repository

Find the latest version and contribute at: https://github.com/codemonkeying/multi-provider-external

📝 License

MIT License - Feel free to modify and distribute!

---

Built by codemonkeying | Version 1.2.1