One API. Every LLM.
A unified gateway to all LLM providers. Smart routing, automatic failover, and built-in optimizations—without changing your code.
Simplified architecture
Replace complex integrations with a single, intelligent endpoint.
Your Application
Single API endpoint
Unified response format
Metriqual Gateway
• Intelligent routing
• Automatic failover
• Cost optimization
• Quality monitoring
30+ Providers
All major LLMs
Custom endpoints
Enterprise-grade gateway features
Smart Routing
Automatically route requests to the best model based on cost, quality, and latency requirements.
Automatic Retries
Built-in retry logic with exponential backoff. Never lose a request to transient errors.
Load Balancing
Distribute load across multiple providers. Avoid rate limits and maximize throughput.
Fallback Models
Define fallback chains. If GPT-4 fails, automatically try Claude or another model.
Rate Limiting
Control costs with built-in rate limiting. Set limits per user, API key, or globally.
Global Edge Network
Route requests through the nearest edge location for minimal latency worldwide.
Provider status
Real-time status of all supported LLM providers
OpenAI
GPT-4, GPT-3.5, DALL-E
Anthropic
Claude 3 Opus, Claude 3 Sonnet
Gemini Pro, PaLM 2
Mistral
Mixtral, Mistral 7B
Cohere
Command, Embed
Meta
Llama 2, CodeLlama
Simple integration
Use any LLM provider with the same clean API
// One endpoint for all providers
const response = await fetch('https://api.metriqual.com/v1/chat/completions', {
headers: {
'Authorization': 'Bearer YOUR_API_KEY',
'Content-Type': 'application/json'
},
body: JSON.stringify({
// Just change the model to switch providers
model: 'gpt-4', // OpenAI
// model: 'claude-3', // Anthropic
// model: 'gemini-pro', // Google
// model: 'mixtral', // Mistral
messages: [{
role: 'user',
content: 'Hello!'
}],
// Optional: Let us pick the best model
routing: {
mode: 'auto',
optimize_for: 'quality', // or 'cost' or 'speed'
max_cost: 0.10
}
})
})
// Consistent response format across all providers
const data = await response.json()
console.log(data.choices[0].message.content)Simplify your LLM infrastructure
Stop managing multiple SDKs and endpoints. Start building with a unified API.