AI Fusion aggregates free-tier API access across Groq, Cerebras, OpenRouter, Gemini, and more โ behind a single OpenAI-compatible endpoint. Drop-in replacement, automatic failover, live progress streaming.
Point any OpenAI-style client at https://ai.viktorarsov.com/v1. That's the entire migration.
# Stream a chat completion across the spiral curl https://ai.viktorarsov.com/v1/chat/completions \ -H "Authorization: Bearer afp_live_โฆ" \ -H "Content-Type: application/json" \ -d '{ "model": "gateway:fast-free", "messages": [{"role": "user", "content": "Hello, fusion."}] }'
from openai import OpenAI client = OpenAI( base_url="https://ai.viktorarsov.com/v1", api_key="afp_live_โฆ", ) resp = client.chat.completions.create( model="gateway:fast-free", messages=[{"role": "user", "content": "Hello, fusion."}], ) print(resp.choices[0].message.content)
Three things every serious gateway needs. We ship all three.
Automatic failover across providers and keys. Atomic Redis allocation, daily cooldowns, midnight resets โ never see a 429 again.
SSE updates as the gateway hops keys and retries. Watch attempts, latencies, and which key answered โ in real time.
Bring your own keys per tenant. We meter every request against retail prices and report your savings versus going direct.
Records-only billing today. Free tier never expires.
Free
Perfect for personal projects and getting started.
$0 / mo
Starter
For developers building real applications.
$10 / mo
Pro
For production workloads requiring scale.
$50 / mo