One API.
Every AI Model.
Infyrence routes your requests to the best available model across OpenAI, Anthropic, Google, Meta, and more - with a single, unified API you already know.
Model Directory
Natively integrated
Google Gemini lineup.
Gemini 2.0 Flash
PrimaryNext-gen multimodal model with state-of-the-art speed and 1M token context.
Gemini 2.0 Flash Lite
ActiveCost-optimized variant of Flash for high-throughput workloads.
Gemini 1.5 Pro
ActiveMid-size model optimized for complex reasoning with the longest context window available.
Gemini 1.5 Flash
ActiveFast and cost-efficient model for everyday multimodal tasks.
Gemini 1.5 Flash 8B
ActiveSmallest and fastest Gemini model, ideal for latency-critical applications.
Platform
Built for production,
not just demos.
Hierarchical Rate Limiting
Enforcing combined RPM and TPM quotas across Virtual Keys, specific models, and global organizational rules. Granular control at every level.
Smart Failover Chains
Advanced routing that handles provider outages dynamically via Exponential Backoff + Jitter to eliminate cascading failures across your stack.
Edge-Level SSRF Shield
Built-in network protection actively blocking malicious redirects to internal infrastructure or loopback IP ranges at the edge layer.
Stream-Safe Quotas
Accurate token tracking capturing usageMetadata directly from the final SSE chunk to instantly manage complimentary balance caps.
Integration
Up and running
in 3 lines of code.
Infyrence is fully compatible with the OpenAI SDK. Change one URL and one model string - that's it. No new SDKs to learn, no breaking changes.
- OpenAI-compatible REST API
- Streaming supported (SSE)
- Function calling & tool use
- JSON mode & structured outputs
const response = await fetch("https://api.infyrence.com/v1/chat/completions", {
method: "POST",
headers: {
"Authorization": "Bearer YOUR_INFYRENCE_KEY",
"Content-Type": "application/json"
},
body: JSON.stringify({
model: "gemini-2.0-flash",
messages: [
{ role: "system", content: "You are an optimized assistant running at the edge." },
{ role: "user", content: "Stream live tokens." }
],
stream: true
})
});