LLM Providers

Universal LLM Integration

Connect to any AI provider through our unified API. Secure credential storage, team management, cost optimization, and automatic failover across OpenAI, Anthropic, Google AI, and more.

View Documentation See Examples

Supported Providers

OpenAI

GPT-4, GPT-3.5 Turbo, and DALL-E integration

Chat Completion Embeddings Image Generation

Anthropic

Claude 3 and Claude 2 models with safety focus

Constitutional AI Long Context Safety First

Google AI

Gemini Pro and PaLM models

Multimodal Code Generation

Azure OpenAI

Enterprise OpenAI through Azure

Enterprise Grade Compliance

Ollama

Local model deployment

On-Premises Privacy

Custom APIs

Connect any OpenAI-compatible API

Flexible Self-Hosted

Encrypted Storage

API keys and credentials are encrypted and securely stored with enterprise-grade security.

  • AES-256 encryption at rest
  • Secure key rotation
  • Audit trail logging
  • Zero-knowledge architecture

Team Management

Organize providers by teams with role-based access controls and usage limits.

  • Team-based provider access
  • Usage quotas and limits
  • Spending controls
  • Usage analytics

Cost Optimization

Monitor spending, optimize model usage, and implement cost controls across providers.

  • Real-time cost tracking
  • Model performance metrics
  • Automatic failover
  • Budget alerts

Unified API

Single API interface works across all providers with consistent response formats.

  • Provider abstraction
  • Consistent interfaces
  • Easy switching
  • Standardized responses

Easy Provider Integration

provider_service.py
Python
# Configure multiple providers
fiber_config = {
    "providers": {
        "openai": {
            "api_key": "sk-...",
            "models": ["gpt-4", "gpt-3.5-turbo"],
            "fallback_priority": 1
        },
        "anthropic": {
            "api_key": "sk-ant-...",
            "models": ["claude-3-opus", "claude-3-sonnet"],
            "fallback_priority": 2
        },
        "azure": {
            "api_key": "...",
            "endpoint": "https://your-resource.openai.azure.com/",
            "models": ["gpt-4"],
            "fallback_priority": 3
        }
    }
}

# Use unified API across all providers
async def generate_with_fallback(prompt, model_preference="gpt-4"):
    try:
        # Try primary provider
        response = await llm_service.generate_completion(
            prompt=prompt,
            model=model_preference,
            provider="openai",
            temperature=0.7
        )
    except ProviderError:
        # Automatic fallback to secondary provider
        response = await llm_service.generate_completion(
            prompt=prompt,
            model="claude-3-sonnet",
            provider="anthropic",
            temperature=0.7
        )
    
    return response

Provider Management Dashboard

LLM Providers

Active Providers 4
Monthly Spend $127.50
Total Requests 12.3K

OpenAI

Active • gpt-4, gpt-3.5-turbo
$89.20 8.1K requests
99.9%

Anthropic

Active • claude-3-sonnet
$31.40 3.2K requests
99.7%

Google AI

Configured • gemini-pro
$6.90 1.0K requests
95.2%