Operational AI Routing Infrastructure·OpenAI & Anthropic compatible

One API. Every Model.

A universal, drop-in gateway that routes across Claude, ChatGPT, DeepSeek, and Gemini — with automatic failover, load balancing, and per-key rate limits. Full OpenAI & Anthropic wire parity.

Powered by Vexanode

one request → any model
POST https://api.botliy.online/v1/chat/completions
Authorization: Bearer rly_••••••••

{ "model": "auto", ... }
✓ routedClaudevia best-available key

Routing across every major provider

ClaudeClaude
OpenAIChatGPT
DeepSeekDeepSeek
GeminiGemini
GrokGrok
MetaLlama
MistralMistral
QwenQwen
KimiKimi
CohereCommand
ClaudeClaude
OpenAIChatGPT
DeepSeekDeepSeek
GeminiGemini
GrokGrok
MetaLlama
MistralMistral
QwenQwen
KimiKimi
CohereCommand
Built for production

Everything you need to route with confidence

Automatic Failover

Upstream 429s and 5xx trigger instant hot-swap to a healthy key — no dropped streams.

Load Balancing

Requests spread across pooled provider keys to maximize throughput and headroom.

Per-Key Rate Limits

Set granular RPM and token ceilings per API key, enforced at the edge.

Prompt Caching

Anthropic-style prompt caching passes straight through, cutting cost and latency.

Low-Latency Routing

Lean dispatch keeps the gateway effectively invisible to your application.

Wire-Protocol Parity

Drop-in OpenAI & Anthropic compatibility — point your SDK at Relay and go.

Drop-in compatible

Change one line. Keep your SDK.

Point your existing OpenAI or Anthropic client at Relay's base URL — no rewrites.

quickstart
from openai import OpenAI

client = OpenAI(
    base_url="https://api.botliy.online/v1",
    api_key="rly_your_key",
)

resp = client.chat.completions.create(
    model="auto",  # Relay picks the best available model
    messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)
How it works

Live in under five minutes

1

Create a key

Sign up and generate a Relay API key with $10 in free credits.

2

Swap the base URL

Point your OpenAI or Anthropic SDK at api.botliy.online.

3

Send requests

Use model: "auto" or name a model — Relay routes and load-balances.

4

Monitor & scale

Track usage, latency, and failover in your dashboard in real time.

Pricing

Simple, transparent pricing

Free

$0

Everything you need to start routing.

  • $10 welcome credits
  • 1 API key
  • All providers
  • Community support
Start free
Recommended

Pro

$29/mo

For teams shipping to production.

  • Unlimited keys
  • Higher rate limits
  • Prompt caching
  • Priority failover
  • Email support
Upgrade to Pro

Enterprise

Custom

Dedicated capacity and controls.

  • SSO & audit logs
  • Dedicated pools
  • SLA & uptime guarantees
  • Solutions engineer
Contact sales
Reliability
0.0%availability target

Built for high availability across pooled providers, with automatic failover that hot-swaps to a healthy key to keep streams alive. Live numbers are always on our status page.

0
providers pooled
0
wire protocols
24/7
failover monitoring
Provider healthAll systems operational
ClaudeOperational
ChatGPTOperational
DeepSeekOperational
GeminiOperational
View live status
FAQ

Questions, answered

Start routing in minutes

Create an account, grab an API key, and point your SDK at Relay. $10 in free credits included.